[prev in list] [next in list] [prev in thread] [next in thread]
List: openmrs-dev
Subject: Re: A clarification regarding "ruled out" and possible or questionable diagnoses
From: Burke Mamlin <burke () openmrs ! org>
Date: 2014-03-26 17:45:12
Message-ID: CAKgo1ZaR1k=CFfuk5BjhXYjvVypWwmUn_FeDFoHeF_WCeZ=8_Q () mail ! gmail ! com
[Download RAW message or body]
Presume & Confirmed should be sufficient modifiers for now and can be
easily understood by everyone. I would not expand this to a list of 6+
modifiers before we know whether or not those two modifiers will be widely
reliably filled and/or used.
-Burke
On Wed, Mar 26, 2014 at 12:39 PM, Andrew Kanter <andy_kanter@yahoo.com>wrote:
> Thanks Ellen,
> Perhaps those classifications should be added to the current diagnosis
> certainty options.
>
>
>
>
> *-------------------- Andrew S. Kanter, MD MPH FACMI*
> Asst. Prof. of Clinical Biomedical Informatics and Clinical Epidemiology
> Columbia University
> Email: andrew.kanter@dbmi.columbia.edu
> Mobile: +1 (646) 469-2421
> Office: +1 (212) 305-4842
> Skype: akanter-ippnw
> Yahoo: andy_kanter
>
>
> On Wednesday, March 26, 2014 4:44 AM, Ellen Ball <ellnball@gmail.com>
> wrote:
>
> My colleague Julie Jensen in Rwanda was explaining that Oncology
> currently differentiates the following:
>
> - suspected - a type of cancer is suspected, but further testing needs
> to be done to determine if it is or is not cancer
> - clinically confirmed - true confirmation can only be done
> pathologically, but there are 2 cancers (KS and cervical) where
> "confirmation" can be done with acceptable accuracy based on visual
> inspection and physical exam (NOTE: there was a lot of discussion among
> the clinicians about the title here. There isn't a clear label for this
> type of confirmation. It's considered confirmation, but not "true"
> confirmation, if that makes any sense!)
> - pathologically confirmed - cancer is only considered truly confirmed
> if pathology is done and indicates cancer cells
>
> Is it useful to add these to the conversation?
>
> Regards,
>
> Ellen Ball
> Partners In Health
>
>
> On Monday, March 24, 2014 3:21:12 PM UTC-4, Jonathan Teich wrote:
>
> I think this is much clearer, and will serve us well for problem lists as
> well as for analyses and decision support.
>
> JT
>
> On Mar 24, 2014, at 2:17 PM, Burke Mamlin <bu...@openmrs.org> wrote:
>
> Thanks, Andy. I think you've given us a path...
>
>
> 1. Diagnoses enterred into the diagnoses list (per visit) and/or
> problem list (across visits) should represent diagnoses the patient has or
> is presumed to have.
>
> 2. If we need to capture "Rule Out Pneumonia", we can create a
> separate "differential diagnosis" list that is explictly understood to
> capture diagnoses to be considered, but that the patient is not presumed to
> have. In most cases, a combination of recording symptoms and/or using the
> "Presumed" modifier can satisfy folks without the need this list.
>
> 3. If we need to capture "Ruled Out Pneumonia", we can create a
> separated "ruled out diagnoses" list that is explicitly understook to
> catpure diagnoses that the patient does *not* have. In most cases,
> people are satisfied to take the absence of a diagnosis and/or its removal
> from a problem list as a proxy for "Ruled Out", making this list only
> needed when the absence of diagnoses needs to be explictly recorded.
>
> While making these differences explicit under the hood (in API and
> database) is important to avoid someone misinterpreting a diagnosis list if
> they don't have or use the modifiers, a client could simplify the
> presentation of these lists for users, even presenting a UI that puts them
> into a single list and lets the user record "Rule Out" and "Ruled Out"
> modifiers, as long as the data are stored in explictly different ways under
> the hood.
>
> -Burke
>
>
>
>
> On Mon, Mar 24, 2014 at 1:50 PM, Andrew Kanter <andy_...@yahoo.com> wrote:
>
> Yikes, quite a good discussion. I would agree with Burke on the
> practicality of adding modifiers like this to typical analytics. I also
> think that the "rule-out" and I suppose "ruled-out" modifiers are really
> hold overs from proper documentation and decision support tools.
>
> Rule-out Diagnosis really means "Pt has signs and symptoms X" and has a
> differential diagnosis of "Y" and Z". The problem list should have the
> signs/symptoms (e.g, Productive Cough) and Possible Pneumonia vs. Possible
> CHF. Analytics/billing are typically done on Productive Cough and the
> differential is used to group tests/workup.
>
> When the CHF is "ruled-out" and Pneumonia is confirmed, the diagnosis is
> changed from productive cough -> pneumonia. There is a link between them.
> Only Productive Cough and Pneumonia are seen on the problem list. In the
> case of differentials, CHF drops off in favor of Pneumonia and this has the
> status of "ruled-out", but it never actually gets that status as the fact
> that it was possible and then replaced with confirmed something else means
> it was ruled-out.
>
> I guess the cough is not really ruled-out as it was certain to start as a
> symptom, but then was upgraded to pneumonia, so you have to distinguish
> between updated problems and ruled-out differential diagnoses.
>
> I DEFINITELY would not want to add "ruled-out" as a diagnosis qualifier if
> we can avoid it.
>
> Andy
>
>
>
>
> *-------------------- Andrew S. Kanter, MD MPH FACMI*
> Asst. Prof. of Clinical Biomedical Informatics and Clinical Epidemiology
> Columbia University
> Email: andrew...@dbmi.columbia. edu
> Mobile: +1 (646) 469-2421
> Office: +1 (212) 305-4842
> Skype: akanter-ippnw
> Yahoo: andy_kanter
>
>
> On Monday, March 24, 2014 12:26 PM, Burke Mamlin <bma...@regenstrief.org>
> wrote:
>
> Sushmitha,
>
> To be clear, which one of these cases were you needing?
>
> - *Ruled Out* Pneumonia – modifier is applied *after* a diagnsosis has
> been considered, investigated, and proven not to be the patient's
> diagnosis. For example, a patient was given the diagnosis of Pneumonia and
> then it is later learned that they do not have pneumonia; rather, they have
> systemic lupus erythematosus.
>
> - *Rule Out* Pneumonia – the clinician enters the modifier + diagnosis
> *before* a diagnosis is made – i.e., when a diagnosis is being
> investigated. For example, a patient presents with one week of dyspnea and
> fever and the chest x-ray has been ordered but not yet done.
>
> -Burke
>
>
> On Mon, Mar 24, 2014 at 11:49 AM, Jonathan Teich <jmt...@gmail.com> wrote:
>
> My concern here is that "ruled out" is unusual in this usage -- so unusual
> that I am wondering if both the phrase and the meaning were slightly
> switched from what was originally clinically intended, during the Bahmni
> discussions.
>
> I did spot the "ruled" vs. "rule" and what it is supposed to mean IF
> everyone notices the difference -- but the difference is small, and
> non-obvious to non-clinicians, so I'm trying to address potential
> confusion.
>
> (I agree with the rest of your note).
>
> Thanks,
> Jonathan
>
>
> On Mar 24, 2014, at 11:30 AM, Burke Mamlin <bma...@regenstrief.org> wrote:
>
> Jonathan, the Bahmni group is considering "Ruled Out" (note the "d"),
> which is the opposite of Confirmed. I agree that clinicians often want to
> work with "Rule Out" (note the lack of "d") with diagnoses and there is a
> continuous battle between the clinical utility of a "Rule Out" modifier and
> the problems it causes for other uses like billing & research. For
> example, many payors will not reimburse for a "Rule Out" diagnosis and
> researchers looking for cohorts by diagnosis can be tricked into including
> patients who do not have the diagnosis if they do not have access &
> understanding of available modifiers. The compromise was to use
> "Presumed", which is semantically different from "Rule Out" (requiring a
> little more commitment to the diagnosis) to try to please the fewest number
> of people. ;-)
>
> If we could be assured of a water-tight control of diagnoses and could
> ensure that it was impossible to get access to a diagnosis without the
> associated modifier, then we could be more confident in the modifiers and
> support not-really-a-diagnosis & negation modifiers like "Rule Out" &
> "Ruled Out". Unfortunately, diagnoses management (like life) is messy and
> we know that the only way that ensuring modifiers accompany diagnoses 100%
> of the time is to pre-coordinate them (i.e., make concepts like "Rule Out
> Pneumonia"), but the explosion of related concepts (at least 3-5 for every
> diagnosis) an even worse mess... I fear even mentioning that approach could
> give Andy angina. So, we compromise by saying to clinicians: "we will give
> you some modifiers, but you can only add items to a patient's diagnosis
> list if you believe they have the diagnosis."
>
> -Burke
>
>
> On Mon, Mar 24, 2014 at 11:04 AM, Jonathan Teich <jmt...@gmail.com> wrote:
>
> [Given the last clarification, this comment also applies to diagnoses as
> opposed to problem lists, except for the part about voiding out diagnoses.]
>
> Jonathan
>
> On Mar 24, 2014, at 10:55 AM, Jonathan Teich <jmt...@gmail.com> wrote:
>
> I just wanted to comment on the phrase "ruled out" and other terms
> defining certainty as they are understood clinically, at least in the US.
>
> Oddly enough, when you see the phrase "rule out", as in "rule out
> pneumonia", it doesn't actually mean that the diagnosis has been ruled out.
> Instead, it means that this is a possible diagnosis that *needs* to be
> ruled out. I know, that's not always the greatest use of language, but that
> is clearly what it has come to mean in US medical documentation. If, for
> example, you see a patient and you think they might have tuberculosis but
> you're not sure and therefore you plan to do more evaluation and testing,
> you may create a problem list entry of "rule out tuberculosis".
>
> Essentially, the phrase "rule out pneumonia" (or any other diagnosis) has
> come to mean pretty much the same thing as "possible pneumonia" or
> "question of pneumonia". I don't see much if any usage of "ruled out" in
> the US as a status modifier to indicate that a diagnosis has been
> eliminated from consideration. In such a case, the problem list item would
> simply be voided or inactivated.
>
> I just wanted to make sure that we didn't hear the phrase "rule out" in
> the above slightly odd but well-used meaning, and translated it into a more
> natural but actually incorrect meaning. If the American usage of "rule
> out" as I've described it is also used worldwide, then we might not want to
> have "ruled out" at all, because of the confusion it would generate.
>
> As a corollary thought -- If you are keeping a standard list of diagnosis
> certainty modifiers, then along with Presumed and Certain you might want to
> include Possible, or something like that.
>
>
> Jonathan
>
> On Mar 24, 2014, at 3:03 AM, Sushmitha S Rao <sus...@thoughtworks.com>
> wrote:
>
> Hi Darius, Burke,
>
> We agree to not save 'ruled out' diagnosis as part of diagnosis
> observation grouping. As you suggested, we will save ruled out diagnosis as
> an observation against a concept : "Ruled out diagnosis". We will be
> sending these 'ruled out' diagnoses as observations in encounter
> transaction.
>
> Also we will not void the original diagnosis that was ruled out. Hence for
> a user flow of a patient for whom a presumed diagnosis is made for
> tuberculosis and then the diagnosis has been ruled out, the observations in
> the database would look as follow :
>
> Obs1
> -RuledOutDiagnosis
> -Tuberculosis
>
> Obs2
> -Diagnosis
> -Tuberculosis
> -Presumed
> -Primary
>
>
>
>
> On Sat, Mar 22, 2014 at 12:10 AM, Jonathan Teich <jte...@gmail.com> wrote:
>
> Very much agree.
>
> Jonathan
>
> *From:* Burke Mamlin [mailto:bma...@regenstrief. org]
> *Sent:* Friday, March 21, 2014 11:52 AM
> *To:* dev-refapp
> *Cc:* dev; Jonathan Teich; Mujiruddin Shaikh; Mário Areias; bahmni
> *Subject:* Re: Extending diagnosis certainty in EMR-API to include more
> possible answers
>
> We have to be *very* careful about modifying diagnoses to the negative –
> i.e., "Ruled out" – since most of the applications & users presume that a
> list of diagnoses are a list of diagnoses and not a list of some thing the
> patient has and some things that the patient does not have. It would
> probably be safer to explictly store "ruled out" diagnoses separately and
> leave it up to applications/clients when these can be safely combined.
>
> This is similar to an allergy list like:
>
> - patient is allergic to sulfa
> - patient is not allergic to penicillins
>
> or a drug list like:
>
> - patient is taking ampicillin/sulbactam
> - patient is not taking warfarin
> - patient taking acetaminophen
>
> There is too great an opportunity for mistakenly harming patients by
> casually mixing positives and negatives.
>
> So, instead of allowing a negation modifier on diagnoses, we could, under
> the hood, create a separate list of diagnoses that have been ruled out.
> Then it would be up to the particular client/application/screen to decide
> whether or not these can be safely presented to the user as a single list
> (even as a "Ruled out" modifier in a diagnosis list). Ensuring that these
> are separate in the model & API will ensure that nobody ever builds a
> cohort of people with diabetes only to discover that many of the patients
> don't have diabetes because their diabetes diagnosis was negated, nobody
> ever mistakenly sends a diagnosis of HIV to a patient's insurer because
> they simply included the diagnosis names and left off the modifiers, etc.
>
> -Burke
>
> On Fri, Mar 21, 2014 at 11:40 AM, Darius Jazayeri <djaz...@gmail.com>
> wrote:
> @Andy Kanter, note this conversation
>
> Hi Sushmitha and Mujir,
>
> That use case makes sense, though I'm not sure that the existing Diagnosis
> mechanism is the way to do it. My concerns:
>
> - currently you can determine if any suspected/confirmed diagnoses
> were made by a simple query on the obs table (and there are some service
> methods and reporting module mechanisms for this), whereas if we made this
> change, then every query for diagnoses would need to join the obs table
> against itself to see if it's a positive or negative diagnosis.
> - Diagnosis Order is meaningless for a diagnosis that is ruled out, so
> I don't think that Diagnosis is the right construct.
>
> My initial thought would be to have another concept like "RULED OUT CODED
> DIAGNOSIS", which doesn't have to belong to an obs group.
>
> This also highlights that we really need to get working on Problem Lists,
> which are the way we ultimately want to manage the evolution of suspected
> -> confirmed or suspected -> ruled out diagnoses over time.
>
> @Andy, @Burke, @Jonathan, thoughts on this? We should definitely discuss
> this on a design call.
>
> -Darius
>
> On Fri, Mar 21, 2014 at 4:03 AM, Sushmitha S Rao <sus...@thoughtworks.com>
> wrote:
> Hi,
>
> We have a requirement in Bahmni to capture certainty for a diagnosis as
> "ruled out". The possible values that we would capture for certainty would
> hence be : Presumed, Confirmed, RuledOut.
>
> From the openMRS Admin UI, we have the ability to add another coded answer
> (e.g. 'Ruled Out') to 'Certainty' concept. However in EMR-API, we
> have org.openmrs.module. emrapi.diagnosis.Diagnosis. Certainty as an enum
> that has possible values of presumed and confirmed.
>
> We would like to change org.openmrs.module. emrapi.diagnosis.Diagnosis.
> Certainty to a class which parses the codes for certainty
> (PRESUMED/CONFIRMED/RULEDOUT) by looking up the set members of the concept
> which is mapped to "Diagnosis Certainty".
>
> This would give us the ability to extend this concept to have any number
> of certainty values. We could further extend this change
> for org.openmrs.module.emrapi. diagnosis.Diagnosis.Order as well.
> This may potentially affect Mirebalais too. Do let us know if you have any
> concerns about this.
>
>
>
--
OpenMRS Developers: http://go.openmrs.org/dev
Post: dev@openmrs.org | Unsubscribe: dev+unsubscribe@openmrs.org
Manage your OpenMRS subscriptions at https://id.openmrs.org/
To unsubscribe from this group and stop receiving emails from it, send an email to \
dev+unsubscribe@openmrs.org.
[Attachment #3 (text/html)]
<div dir="ltr"><div>Presume & Confirmed should be sufficient modifiers for now \
and can be easily understood by everyone. I would not expand this to a list of 6+ \
modifiers before we know whether or not those two modifiers will be widely reliably \
filled and/or used.</div>
<div><br></div><div class="gmail_extra">-Burke<br><br><div class="gmail_quote">
On Wed, Mar 26, 2014 at 12:39 PM, Andrew Kanter <span dir="ltr"><<a \
href="mailto:andy_kanter@yahoo.com" \
target="_blank">andy_kanter@yahoo.com</a>></span> wrote:<br> <blockquote \
class="gmail_quote" style="margin:0px 0px 0px \
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div><div \
style="font-size:12pt;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida Grande',sans-serif">
<div>
<span>Thanks Ellen,</span></div><div \
style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida Grande',sans-serif"><span>Perhaps those \
classifications should be added to the current diagnosis certainty \
options.</span></div>
<div><div></div><div> </div><div><b>-------------------- <br>Andrew S. Kanter, MD \
MPH FACMI<br><span style="font-weight:normal"><b><br></b></span></b></div><div \
style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,clean,sans-serif">
Asst. Prof. of Clinical Biomedical Informatics and Clinical \
Epidemiology<br></div></div><div \
style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,clean,sans-serif">Columbia \
University<br>
Email: <a href="mailto:andrew.kanter@dbmi.columbia.edu" \
target="_blank">andrew.kanter@dbmi.columbia.edu</a> <br><div>Mobile: <a \
href="tel:%2B1%20%28646%29%20469-2421" value="+16464692421" target="_blank">+1 (646) \
469-2421</a><br>
Office: <a href="tel:%2B1%20%28212%29%20305-4842" value="+12123054842" \
target="_blank">+1 (212) 305-4842</a><br>Skype: akanter-ippnw<br>Yahoo: \
andy_kanter</div></div><div style="display:block"> <br> <br> <div \
style="font-family:HelveticaNeue,'Helvetica Neue',Helvetica,Arial,'Lucida \
Grande',sans-serif;font-size:12pt">
<div style="font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida \
Grande',sans-serif;font-size:12pt"><div><div> <div dir="ltr"> <font face="Arial"> \
On Wednesday, March 26, 2014 4:44 AM, Ellen Ball <<a \
href="mailto:ellnball@gmail.com" target="_blank">ellnball@gmail.com</a>> \
wrote:<br>
</font> </div> </div></div><blockquote \
style="border-left-width:2px;border-left-style:solid;border-left-color:rgb(16,16,255);margin-left:5px;margin-top:5px;padding-left:5px"> \
<div><div><div dir="ltr"><div><div>My colleague Julie Jensen in Rwanda was explaining \
that <span style="font-size:13px">Oncology currently differentiates the \
following:</span><div><div><div><ul><li><span style="font-size:13px">suspected - a \
type of cancer is suspected, but further testing needs to be done to determine if it \
is or is not cancer</span></li>
<li><span style="font-size:13px">clinically confirmed - true confirmation can only be \
done pathologically, but there are 2 cancers (KS and cervical) where
"confirmation" can be done with acceptable accuracy based on visual
inspection and physical exam (NOTE: there was a lot of discussion
among the clinicians about the title here. There isn't a clear label
for this type of confirmation. It's considered confirmation, but not
"true" confirmation, if that makes any sense!)</span></li><li><span \
style="font-size:13px">pathologically confirmed - cancer is only considered truly \
confirmed if pathology is done and indicates cancer cells</span><br>
</li></ul>
</div>
</div></div><div>Is it useful to add these to the \
conversation?</div><div><br></div><div>Regards,</div><div><br></div><div>Ellen \
Ball</div><div>Partners In Health</div></div></div><div><div><div><br><br>On Monday, \
March 24, 2014 3:21:12 PM UTC-4, Jonathan Teich wrote:</div>
</div><blockquote style="margin:0px 0px 0px \
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div><div><div><div>I \
think this is much clearer, and will serve us well for problem lists as well as for \
analyses and decision support. <br>
<br>JT</div><div><br>On Mar 24, 2014, at 2:17 PM, Burke Mamlin <<a \
rel="nofollow">bu...@openmrs.org</a>> wrote:<br><br></div></div></div><blockquote \
type="cite"><div><div><div><div dir="ltr">Thanks, Andy. I think you've given us \
a path...<div>
<br></div><div><ol><li>Diagnoses enterred into the diagnoses list (per visit) and/or \
problem list (across visits) should represent diagnoses the patient has or is \
presumed to have.<br>
</li><li>If we need to capture "Rule Out Pneumonia", we can create a \
separate "differential diagnosis" list that is explictly understood to \
capture diagnoses to be considered, but that the patient is not presumed to have. \
In most cases, a combination of recording symptoms and/or using the \
"Presumed" modifier can satisfy folks without the need this list.<br>
</li><li>If we need to capture "Ruled Out Pneumonia", we can create a \
separated "ruled out diagnoses" list that is explicitly understook to \
catpure diagnoses that the patient does <i>not</i> have. In most cases, people are \
satisfied to take the absence of a diagnosis and/or its removal from a problem list \
as a proxy for "Ruled Out", making this list only needed when the absence \
of diagnoses needs to be explictly recorded.</li>
</ol></div><div>While making these differences explicit under the hood (in API and \
database) is important to avoid someone misinterpreting a diagnosis list if they \
don't have or use the modifiers, a client could simplify the presentation of \
these lists for users, even presenting a UI that puts them into a single list and \
lets the user record "Rule Out" and "Ruled Out" modifiers, as \
long as the data are stored in explictly different ways under the hood.<br>
</div><div><br></div><div>-Burke</div><div><br></div><div><br></div></div></div></div><div><br><br><div><div><div>On \
Mon, Mar 24, 2014 at 1:50 PM, Andrew Kanter <span dir="ltr"><<a \
rel="nofollow">andy_...@yahoo.com</a>></span> wrote:<br>
</div></div><blockquote style="margin:0px 0px 0px \
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div><div \
style="font-size:12pt;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida Grande',sans-serif">
<div><div><div>
<span>Yikes, quite a good discussion. I would agree with Burke on the practicality of \
adding modifiers like this to typical analytics. I also think that the \
"rule-out" and I suppose "ruled-out" modifiers are really hold \
overs from proper documentation and decision support tools.</span></div>
<div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida \
Grande',sans-serif"><span><br></span></div><div \
style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida Grande',sans-serif">
<span>Rule-out Diagnosis really means "Pt has signs and symptoms X" and has \
a differential diagnosis of "Y" and Z". The problem list should have \
the signs/symptoms (e.g, Productive Cough) and Possible Pneumonia vs. Possible CHF. \
Analytics/billing are typically done on Productive Cough and the differential is used \
to group tests/workup.</span></div>
<div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida \
Grande',sans-serif"><span><br></span></div><div \
style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida Grande',sans-serif">
<span>When the CHF is "ruled-out" and Pneumonia is confirmed, the diagnosis \
is changed from productive cough -> pneumonia. There is a link between them. Only \
Productive Cough and Pneumonia are seen on the problem list. In the case of \
differentials, CHF drops off in favor of Pneumonia and this has the status of \
"ruled-out", but it never actually gets that status as the fact that it was \
possible and then replaced with confirmed something else means it was ruled-out. \
</span></div><div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida Grande',sans-serif">
<span><br></span></div><div \
style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida Grande',sans-serif"><span>I guess the cough \
is not really ruled-out as it was certain to start as a symptom, but then was \
upgraded to pneumonia, so you have to distinguish between updated problems and \
ruled-out differential diagnoses.</span></div>
<div><span><br></span></div><div \
style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida Grande',sans-serif"><span>I DEFINITELY \
would not want to add "ruled-out" as a diagnosis qualifier if we can avoid \
it.</span></div>
<div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida \
Grande',sans-serif"><span><br></span></div><div \
style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida Grande',sans-serif">
<span>Andy</span></div><div></div><div> </div><div><b>-------------------- \
<br>Andrew S. Kanter, MD MPH FACMI<br><span \
style="font-weight:normal"><b><br></b></span></b></div><div \
style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,clean,sans-serif">
Asst. Prof. of Clinical Biomedical Informatics and Clinical \
Epidemiology<br></div><div \
style="font-style:normal;font-size:13px;background-color:transparent;font-family:arial,helvetica,clean,sans-serif">Columbia \
University<br>
Email: <a rel="nofollow">andrew...@dbmi.columbia. edu</a> <br>Mobile: <a \
rel="nofollow">+1 (646) 469-2421</a><br>
Office: <a rel="nofollow">+1 (212) 305-4842</a><br>Skype: akanter-ippnw<br>Yahoo: \
andy_kanter</div></div></div><div style="display:block"> <br> <br> <div \
style="font-family:HelveticaNeue,'Helvetica Neue',Helvetica,Arial,'Lucida \
Grande',sans-serif;font-size:12pt">
<div style="font-family:HelveticaNeue,'Helvetica \
Neue',Helvetica,Arial,'Lucida \
Grande',sans-serif;font-size:12pt"><div><div><div><div> <div dir="ltr"> <font \
face="Arial"> On Monday, March 24, 2014 12:26 PM, Burke Mamlin <<a \
rel="nofollow">bma...@regenstrief.org</a>> wrote:<br> </font> </div> \
</div></div></div></div><blockquote \
style="border-left-width:2px;border-left-style:solid;border-left-color:rgb(16,16,255);margin-left:5px;margin-top:5px;padding-left:5px">
<div><div><div><div><div><div><div><div dir="ltr">Sushmitha,<br \
clear="none"><div><br clear="none"></div><div>To be clear, which one of these cases \
were you needing?</div><div><ul><li><b>Ruled Out</b> Pneumonia – modifier is \
applied <i>after</i> a diagnsosis has been considered, investigated, and proven not \
to be the patient's diagnosis. For example, a patient was given the diagnosis \
of Pneumonia and then it is later learned that they do not have pneumonia; rather, \
they have systemic lupus erythematosus.<br clear="none">
</li><li><b>Rule Out</b> Pneumonia – the clinician enters the modifier + \
diagnosis <i>before</i> a diagnosis is made – i.e., when a diagnosis is being \
investigated. For example, a patient presents with one week of dyspnea and fever \
and the chest x-ray has been ordered but not yet done.</li>
</ul><div>-Burke</div></div></div></div></div><div><br clear="none"><br \
clear="none"><div><div><div>On Mon, Mar 24, 2014 at 11:49 AM, Jonathan Teich <span \
dir="ltr"><<a rel="nofollow" shape="rect">jmt...@gmail.com</a>></span> \
wrote:<br clear="none">
</div></div><blockquote style="margin:0px 0px 0px \
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div><div><div><div>My \
concern here is that "ruled out" is unusual in this usage -- so unusual \
that I am wondering if both the phrase and the meaning were slightly switched from \
what was originally clinically intended, during the Bahmni discussions. </div>
<div><br clear="none"></div><div>I did spot the "ruled" vs. \
"rule" and what it is supposed to mean IF everyone notices the difference \
-- but the difference is small, and non-obvious to non-clinicians, so I'm trying \
to address potential confusion. </div>
<div><br clear="none"></div><div>(I agree with the rest of your note).<br \
clear="none"><br clear="none">Thanks,</div><div>Jonathan</div></div></div><div><div><div><div><div><br \
clear="none"></div><div><br clear="none"> On Mar 24, 2014, at 11:30 AM, Burke Mamlin \
<<a rel="nofollow" shape="rect">bma...@regenstrief.org</a>> wrote:<br \
clear="none">
<br clear="none"></div></div></div><blockquote type="cite"><div><div><div><div \
dir="ltr">Jonathan, the Bahmni group is considering "Ruled Out" (note the \
"d"), which is the opposite of Confirmed. I agree that clinicians often \
want to work with "Rule Out" (note the lack of "d") with \
diagnoses and there is a continuous battle between the clinical utility of a \
"Rule Out" modifier and the problems it causes for other uses like billing \
& research. For example, many payors will not reimburse for a "Rule \
Out" diagnosis and researchers looking for cohorts by diagnosis can be tricked \
into including patients who do not have the diagnosis if they do not have access \
& understanding of available modifiers. The compromise was to use \
"Presumed", which is semantically different from "Rule Out" \
(requiring a little more commitment to the diagnosis) to try to please the fewest \
number of people. ;-)<div>
<br clear="none"></div><div>If we could be assured of a water-tight control of \
diagnoses and could ensure that it was impossible to get access to a diagnosis \
without the associated modifier, then we could be more confident in the modifiers and \
support not-really-a-diagnosis & negation modifiers like "Rule Out" \
& "Ruled Out". Unfortunately, diagnoses management (like life) is \
messy and we know that the only way that ensuring modifiers accompany diagnoses 100% \
of the time is to pre-coordinate them (i.e., make concepts like "Rule Out \
Pneumonia"), but the explosion of related concepts (at least 3-5 for every \
diagnosis) an even worse mess... I fear even mentioning that approach could give Andy \
angina. So, we compromise by saying to clinicians: "we will give you some \
modifiers, but you can only add items to a patient's diagnosis list if you \
believe they have the diagnosis."</div>
<div><br clear="none"></div><div>-Burke</div></div></div></div><div><br \
clear="none"><br clear="none"><div><div><div>On Mon, Mar 24, 2014 at 11:04 AM, \
Jonathan Teich <span dir="ltr"><<a rel="nofollow" \
shape="rect">jmt...@gmail.com</a>></span> wrote:<br clear="none">
</div></div><blockquote style="margin:0px 0px 0px \
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div><div><div><div>[Given \
the last clarification, this comment also applies to diagnoses as opposed to problem \
lists, except for the part about voiding out diagnoses.]<span><font \
color="#888888"><br clear="none">
<br clear="none">Jonathan</font></span></div></div></div><div><div><div><div><div><br \
clear="none">On Mar 24, 2014, at 10:55 AM, Jonathan Teich <<a rel="nofollow" \
shape="rect">jmt...@gmail.com</a>> wrote:<br clear="none">
<br clear="none"></div></div></div><blockquote type="cite">
<div><div><div><div>I just wanted to comment on the phrase "ruled out" and \
other terms defining certainty as they are understood clinically, at least in the US. \
<br clear="none"><br clear="none">Oddly enough, when you see the phrase "rule \
out", as in "rule out pneumonia", it doesn't actually mean that \
the diagnosis has been ruled out. Instead, it means that this is a possible diagnosis \
that *needs* to be ruled out. I know, that's not always the greatest use of \
language, but that is clearly what it has come to mean in US medical documentation. \
If, for example, you see a patient and you think they might have tuberculosis but \
you're not sure and therefore you plan to do more evaluation and testing, you may \
create a problem list entry of "rule out tuberculosis". </div>
<div><br clear="none"></div><div>Essentially, the phrase "rule out \
pneumonia" (or any other diagnosis) has come to mean pretty much the same thing \
as "possible pneumonia" or "question of pneumonia". I don't \
see much if any usage of "ruled out" in the US as a status modifier to \
indicate that a diagnosis has been eliminated from consideration. In such a case, \
the problem list item would simply be voided or inactivated.</div>
<div><br clear="none"></div><div>I just wanted to make sure that we didn't hear \
the phrase "rule out" in the above slightly odd but well-used meaning, and \
translated it into a more natural but actually incorrect meaning. If the American \
usage of "rule out" as I've described it is also used worldwide, then \
we might not want to have "ruled out" at all, because of the confusion it \
would generate.</div>
<div><br clear="none"></div><div>As a corollary thought -- If you are keeping a \
standard list of diagnosis certainty modifiers, then along with Presumed and Certain \
you might want to include Possible, or something like that.<br clear="none">
<br clear="none"><br clear="none">
Jonathan</div><div><br clear="none">On Mar 24, 2014, at 3:03 AM, Sushmitha S Rao \
<<a rel="nofollow" shape="rect">sus...@thoughtworks.com</a>> wrote:<br \
clear="none">
<br clear="none"></div></div></div><blockquote type="cite"><div><div><div><div \
dir="ltr">
Hi Darius, Burke,<div><br clear="none"></div><div>We agree to not save 'ruled \
out' diagnosis as part of diagnosis observation grouping. As you suggested, we \
will save ruled out diagnosis as an observation against a concept : "Ruled out \
diagnosis". We will be sending these 'ruled out' diagnoses as \
observations in encounter transaction. </div>
<div><br clear="none"></div><div>Also we will not void the original diagnosis that \
was ruled out. Hence for a user flow of a patient for whom a presumed diagnosis is \
made for tuberculosis and then the diagnosis has been ruled out, the observations in \
the database would look as follow : </div>
<div><br clear="none"></div><div>Obs1</div><div>-RuledOutDiagnosis</div><div>-Tuberculosis</div><div><br \
clear="none"></div><div>Obs2</div><div>-Diagnosis</div><div> \
-Tuberculosis</div><div> -Presumed</div><div> -Primary</div>
<div><br clear="none"></div><div>
<br clear="none"></div></div></div></div><div><br clear="none"><br \
clear="none"><div><div><div>On Sat, Mar 22, 2014 at 12:10 AM, Jonathan Teich <span \
dir="ltr"><<a rel="nofollow" shape="rect">jte...@gmail.com</a>></span> \
wrote:<br clear="none">
</div></div><blockquote style="margin:0px 0px 0px \
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div \
lang="EN-US"><div><div><div><div><span style="font-size:11pt">Very much \
agree.<u></u><u></u></span></div>
<div><span style="font-size:11pt"><u></u> <u></u></span></div><div><span \
style="font-size:11pt">Jonathan<u></u><u></u></span></div>
<div><span style="font-size:11pt"><u></u> <u></u></span></div><div><b><span \
style="font-size:10pt">From:</span></b><span style="font-size:10pt"> Burke Mamlin \
[mailto:<a rel="nofollow" shape="rect">bma...@regenstrief. org</a>] <br clear="none">
<b>Sent:</b> Friday, March 21, 2014 11:52 AM<br clear="none"><b>To:</b> dev-refapp<br \
clear="none"><b>Cc:</b> dev; Jonathan Teich; Mujiruddin Shaikh; Mário Areias; \
bahmni<br clear="none"><b>Subject:</b> Re: Extending diagnosis certainty in EMR-API \
to include more possible answers<u></u><u></u></span></div>
</div></div><div><div><div><div><div><u></u> <u></u></div><div><div>We have to be \
<i>very</i> careful about modifying diagnoses to the negative – i.e., "Ruled \
out" – since most of the applications & users presume that a list of \
diagnoses are a list of diagnoses and not a list of some thing the patient has and \
some things that the patient does not have. It would probably be safer to explictly \
store "ruled out" diagnoses separately and leave it up to \
applications/clients when these can be safely combined.<u></u><u></u></div>
<div><div><u></u> <u></u></div></div><div><div>This is similar to an allergy list \
like:<u></u><u></u></div></div><div><ul type="disc"><li>patient is allergic to \
sulfa<u></u><u></u></li><li>patient is not allergic to penicillins<u></u><u></u></li>
</ul><div><div>or a drug list like:<u></u><u></u></div></div></div><div><ul \
type="disc"><li>patient is taking ampicillin/sulbactam<u></u><u></u></li><li>patient \
is not taking warfarin<u></u><u></u></li><li>patient taking \
acetaminophen<u></u><u></u></li>
</ul><div><div>There is too great an opportunity for mistakenly harming patients by \
casually mixing positives and negatives.<u></u><u></u></div>
</div></div><div><div><u></u> <u></u></div></div><div><div>So, instead of allowing a \
negation modifier on diagnoses, we could, under the hood, create a separate list of \
diagnoses that have been ruled out. Then it would be up to the particular \
client/application/screen to decide whether or not these can be safely presented to \
the user as a single list (even as a "Ruled out" modifier in a diagnosis \
list). Ensuring that these are separate in the model & API will ensure that \
nobody ever builds a cohort of people with diabetes only to discover that many of the \
patients don't have diabetes because their diabetes diagnosis was negated, nobody \
ever mistakenly sends a diagnosis of HIV to a patient's insurer because they \
simply included the diagnosis names and left off the modifiers, \
etc.<u></u><u></u></div>
</div><div><div><div><u></u> \
<u></u></div></div><div><div>-Burke<u></u><u></u></div></div></div></div></div></div><div><div \
style="margin-bottom:12pt"><u></u> <u></u></div><div><div><div><div>
On Fri, Mar 21, 2014 at 11:40 AM, Darius Jazayeri <<a rel="nofollow" \
shape="rect">djaz...@gmail.com</a>> wrote:<u></u><u></u></div><div><div>@Andy \
Kanter, note this conversation<u></u><u></u></div>
<div><div><u></u> <u></u></div></div><div><div>Hi Sushmitha and \
Mujir,<u></u><u></u></div></div><div><div><u></u> <u></u></div></div><div><div>That \
use case makes sense, though I'm not sure that the existing Diagnosis mechanism \
is the way to do it. My concerns:<u></u><u></u></div>
</div><div><ul type="disc"><li>currently you can determine if any suspected/confirmed \
diagnoses were made by a simple query on the obs table (and there are some service \
methods and reporting module mechanisms for this), whereas if we made this change, \
then every query for diagnoses would need to join the obs table against itself to see \
if it's a positive or negative diagnosis.<u></u><u></u></li>
<li>Diagnosis Order is meaningless for a diagnosis that is ruled out, so I don't \
think that Diagnosis is the right construct.<u></u><u></u></li></ul><div><div>My \
initial thought would be to have another concept like "RULED OUT CODED \
DIAGNOSIS", which doesn't have to belong to an obs \
group.<u></u><u></u></div>
</div><div><div><u></u> <u></u></div></div><div><div>This also highlights that we \
really need to get working on Problem Lists, which are the way we ultimately want to \
manage the evolution of suspected -> confirmed or suspected -> ruled out \
diagnoses over time.<u></u><u></u></div>
</div></div><div><div><u></u> <u></u></div></div><div><div>@Andy, @Burke, @Jonathan, \
thoughts on this? We should definitely discuss this on a design \
call.<u></u><u></u></div></div><div><div>
<span style="color:rgb(136,136,136)"><u></u> \
<u></u></span></div></div><div><div><span \
style="color:rgb(136,136,136)">-Darius<u></u><u></u></span></div></div></div></div></div><div><div><div><div \
style="margin-bottom:12pt">
<u></u> <u></u></div><div><div><div><div>On Fri, Mar 21, 2014 at 4:03 AM, Sushmitha \
S Rao <<a rel="nofollow" shape="rect">sus...@thoughtworks.com</a>> \
wrote:<u></u><u></u></div>
</div></div><div><div><div><div>
Hi,<u></u><u></u></div><div><div><u></u> <u></u></div></div><div><div>We have a \
requirement in Bahmni to capture certainty for a diagnosis as "ruled out". \
The possible values that we would capture for certainty would hence be : Presumed, \
Confirmed, RuledOut.<u></u><u></u></div>
</div><div><div><u></u> <u></u></div></div></div></div><div><div>From the openMRS \
Admin UI, we have the ability to add another coded answer (e.g. 'Ruled Out') \
to 'Certainty' concept. However in EMR-API, we have org.openmrs.module. \
emrapi.diagnosis.Diagnosis. Certainty as an enum that has possible values of presumed \
and confirmed. <u></u><u></u></div>
</div><div><div><u></u> <u></u></div></div><div><div>We would like to change \
org.openmrs.module. emrapi.diagnosis.Diagnosis. Certainty to a class which parses the \
codes for certainty (PRESUMED/CONFIRMED/RULEDOUT) by looking up the set members of \
the concept which is mapped to "Diagnosis Certainty".<u></u><u></u></div>
</div><div><div><u></u> <u></u></div></div><div><div>This would give us the ability \
to extend this concept to have any number of certainty values. We could further \
extend this change for org.openmrs.module.emrapi. diagnosis.Diagnosis.Order as \
well.<u></u><u></u></div>
</div><div><div><div>This may potentially affect Mirebalais too. Do let us know if \
you have any concerns about this. <span style="color:rgb(136,136,136)"><br \
clear="all"><u></u><u></u></span></div><div><div>
<span style="color:rgb(136,136,136)"><u></u> \
</span></div></div></div></div></div></div></div></div></div></div></div></div></div>< \
/div></div></blockquote></div></div></div></blockquote></div></blockquote></div></div></div>
</blockquote>
</div></div></div></blockquote></div></div></div></blockquote></div></div></div></div> \
</div></div></div></blockquote></div></div></div></div></div></blockquote></div></div></div></blockquote></div></blockquote></div></div>
</div></div></blockquote></div></div></div></div></div></blockquote></div></div></div>
<p></p>
-- <br />
OpenMRS Developers: <a \
href="http://go.openmrs.org/dev">http://go.openmrs.org/dev</a><br />
Post: dev@openmrs.org | Unsubscribe: dev+unsubscribe@openmrs.org<br />
Manage your OpenMRS subscriptions at <a \
href="https://id.openmrs.org/">https://id.openmrs.org/</a><br />
<p></p>
To unsubscribe from this group and stop receiving emails from it, send an email to <a \
href="mailto:dev+unsubscribe@openmrs.org">dev+unsubscribe@openmrs.org</a>.<br />
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic