Fwd: [IMPROVEDX] IOM report is released - Diagnosis in actual practice

Julianne Nemes Walsh nemeswalsh at GMAIL.COM
Thu Oct 8 12:07:00 UTC 2015


Thank you for sharing your tool. I will test it out over the next few weeks
and try to give you some feedback.  As a faculty member for a PNP program I
am always seeking methods to broaden the students mind mapping experiences
so they may have a better understanding of the cognitive process and the
use of probabilities.

Julianne Nemes Walsh

On Tue, Oct 6, 2015 at 9:13 PM, Vipindas Chengat <syncopesystem at gmail.com>
wrote:

> I agree with both view points. :)
>
> As Dr. Jain points out, there's always been a problem of taking a Bayesian
> approach while trying to use epidemiological data. We’d never go to a
> patient's room armed with a logarithmic table of probabilities. Instead, we
> work to understand the relation between symptoms and diagnoses from a
> patho-physiological perspective. We never know what the exact base rate
> is. In the example he gave, 7% is the prevalence of CAD in that population
> but we can't say for sure that it is the same probability for patients who
> come to an ER with acute chest pain.
>
> At the same time, we definitely use probabilities, too, as Dr. Brush
> points out. There is no way we can practice evidence based medicine without
> considering probabilities and likelihood ratios.
>
> Keeping all this in mind, I designed an algorithm (licensed to Physician
> Cognition) to work based on both pattern recognition AND Bayesian
> principles. It is designed to improve its results continuously from both
> user feedback and data analysis. I would like to provide it free of charge
> for any research projects on the issues raised in this thread and at the
> recent DEM conference. With over half a million lines of code, it can map
> almost every relation that is useful to make a clinical diagnosis.
>
> I am uncomfortable talking about my own project here. But since my team
> and I have already done enough heavy lifting to move a mountain, I would be
> doing patients a disservice if I stood by and let a group of expert and
> committed clinicians try to reinvent a wheel that can’t be built without a
> big team over a long time.
>
> The algorithm is available for free at beta.physiciancognition.com.
> Faster, simpler mobile apps will be available in a few weeks. For any
> combination of unlimited numbers of symptoms, signs, labs, medications, and
> histories, the software, after asking crucial questions, lists differential
> diagnoses in the order of relevance and recommends further questions
> designed to narrow down the differentials. By answering a few questions,
> the list and the order changes considerably. It works just the same way as
> we clinicians do, minus the cognitive error and bias. It’s a great tool
> for avoiding under- or over-utilization, as well as for empowering and
> teaching NPs, PAs, and residents.
>
> You will sometimes find less than perfect results from the algorithm.
> There is a feedback video in the top center (after log-in) that explains
> how users can take action that will result in improving the system, for
> everyone, within days or hours. I believe that is how we scalably improve
> both medicine and healthcare globally.
>
>
> *Vipindas Chengat, MD FACP**  |  *Chairman, Physician Cognition, Inc.
>   —————————————————————————————————
>   Mobile: +1 (773) 575-3550
>   Email: Vipin at PhysicianCognition.Com <Vipin at PhysicianCognition.com>
>   Website: PhysicianCognition.Com <http://physiciancognition.com/>
>
>
>
> On Sat, Oct 3, 2015 at 7:58 AM, John Brush <jebrush at me.com> wrote:
>
>> I’m afraid that I can’t agree with Dr. Jain’s argument. I think his
>> argument is circular, difficulty to follow, and selectively self-serving.
>> We have an adage in medicine: “Common things are common.” Otherwise,
>> every diagnostic exercise would become a wild goose chase, leading us to
>> look into every remote possibility every time. Having said that, I can also
>> say that if we collect cases over time, uncommon things become common.
>> Someone somewhere will eventually win the lottery. Uncommon diagnoses do
>> occur eventually. But the exceptions should not define the rules.
>> The STEMI case that Dr. Jain presents proves my point. I am in
>> interventional cardiologist who frequently takes patients with suspected
>> STEMI to the cath lab for intervention. I have been getting direct feedback
>> on these cases for about 25 years. I can tell you that there is a false
>> positive rate of about 15% among STEMI alerts that are taken to the cath
>> lab (numerous reports in the literature confirm that estimate). We allow
>> that false positive rate because we make a subjective calculation of
>> expected value. Even if a patient has a relatively low initial prior
>> probability of STEMI, like Dr. Jain’s example, we don’t want to miss a
>> serious diagnosis like a STEMI. The EKG findings change the probability
>> estimate and make a STEMI quite plausible in such a patient. In a patient
>> like Dr. Jain’s example, we know that there is about a 50-50 chance of
>> finding an occluded artery, which is certainly high enough to activate the
>> cath lab. And sure enough, over time, 50% is about the frequency that we
>> find in such patients.
>> Dr. Jain references central limit theorem. That theorem applies to
>> probability for a continuous variable, and states that for any
>> distribution, the sample means of repeated samples will become a normal
>> distribution. I’m not sure I follow his argument that it applies to a
>> probability distribution of categorial variables. A diagnostic category is
>> a countable variable. Kolmogorov’s principles, however, do apply. The
>> probabilities of all of the possibilities do add up to one, if they are all
>> independent. General knowledge of these probability principles can help us
>> organize our thinking.
>> When we see a patient with chest pain in the ED, we start to narrow the
>> sample space by asking questions and making observations. For example, we
>> can eliminate the possibility of a stab wound very quickly by noticing that
>> there is no knife in the chest. Through early hypothesis generation, we
>> narrow the range of possibilities to the point were we can start the
>> process of iterative hypothesis testing. We have at our disposal many
>> possible tests that we can perform. We can send a troponin, do a CT scan
>> for dissection, do a stress echo, go directly to the cath lab, etc. We
>> can’t do all of these tests at the same time, and we probably don’t want to
>> do every test on every patient. So how do we decide what test to do first?
>> We do a little mental calculation of the subjective probabilities, which
>> gives us an idea of the expected value of each test. We don’t want to miss
>> a diagnosis with serious consequences, like MI or dissection, so an EKG and
>> CXR are done on virtually everyone, regardless of the prior probability.
>> But we narrow the sample space as we hone in on the correct diagnosis. We
>> don’t want to narrow the search prematurely, and we use a differential
>> diagnosis to help us guard against jumping to conclusions. All of this is
>> guided by some notion of relative probabilities.
>> Dr. Jain talks about the CPC method of diagnosis. This is a useful
>> pedagogical exercise, where an expert can expound on clinical medicine, but
>> it is very artificial, as compared to real world practice. An expert may
>> spend days or weeks preparing a CPC discussion. His/her main goals are to
>> not miss the diagnosis, and to eloquently discuss all of the possibilities.
>> It is almost purely System 2 thinking. In the real world, with time
>> constraints and uncertainty, we have to employ System 1’s intuition. It is
>> helpful, however, if we calibrate our intuition through knowledge of the
>> relative strength of evidence and the base rates of various diagnostic
>> possibilities. I think that having an intuitive sense of probability is the
>> essence of experiential knowledge. Savvy clinicians make good bets.
>> The fundamental assumption of evidence based medicine is that the
>> frequencies that we measure in populations of patients can be applied to an
>> individual patient. The measured frequencies from our aggregated
>> experience, or from the reports in the literature inform us on how we
>> should think about an individual. Single event or single patient
>> probability then becomes a degree of belief, which is then modified by
>> additional information that we gain through diagnostic testing. In fact,
>> the sensitivity and specificity of diagnostic tests are defined using a
>> frequency notion of probability. They are cumulative probabilities,
>> depending on where we draw the line of demarcation. Some tests, like an
>> x-ray for a broken arm, are so compelling that they lead to absolute
>> certainty. Other tests, like EKGs, stress tests, troponins, etc, don’t have
>> perfect operating characteristics, however, and we are left with a
>> probability estimate for each diagnostic possibility that is somewhere
>> between 0 and 1. Usually we get to a point of certainty, but sometimes,
>> through adductive reasoning, we are left with the most plausible diagnosis,
>> but never really know for sure.
>> I hate to drag the listserv through this back and forth again, but to me,
>> Dr. Jain’s arguments seem to counter what we have been taught about
>> evidence-based medicine, but also run counter to principles of cognitive
>> psychology. Without some intuitive idea of probability and likelihood, we
>> would be totally adrift in clinical medicine, so I just can’t let this go.
>> John
>>
>> John E. Brush, Jr., M.D., FACC
>> Professor of Medicine
>> Eastern Virginia Medical School
>> Sentara Cardiology Specialists
>> 844 Kempsville Road, Suite 204
>> Norfolk, VA 23502
>> 757-261-0700
>> Cell: 757-477-1990
>> jebrush at me.com
>>
>>
>>
>> On Oct 2, 2015, at 2:45 PM, Mark Graber <mark.graber at IMPROVEDIAGNOSIS.ORG>
>> wrote:
>>
>> Note and manuscript forwarded on behalf of Dr Bimal Jain.
>>
>> ------------------------------
>>
>>
>> To unsubscribe from IMPROVEDX: click the following link:
>>
>> http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1
>>
>> or send email to: IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG
>>
>>
>>
>> Moderator:David Meyers, Board Member, Society for Improving Diagnosis in
>> Medicine
>>
>> To learn more about SIDM visit:
>> http://www.improvediagnosis.org/
>>
>> <Bimal Jain - The Role of Probability in Diagnosis.docx>
>>
>>
>>
>> *From:* Jain, Bimal P.,M.D.
>> *Sent:* Thursday, October 01, 2015 1:54 PM
>> *To:* 'Mark Graber'
>> *Subject:* RE: [IMPROVEDX] IOM report is released - Diagnosis in actual
>> practice
>>
>>
>>
>> Hi Mark and all,
>>
>>
>> It is important to understand how diagnosis is performed in actual
>> practice as a correct diagnosis is made after all  85 percent of the time
>> in practice. To reduce diagnostic errors, we need to know if the method in
>> practice needs to be improved or whether certain deviations from it need to
>> be eliminated. The most puzzling issue in this regard is the role that
>> probability plays or does not play in diagnosis. The puzzle arises because
>> a probabilistic approach has been prescribed for a long time, but it does
>> not appear to be employed in practice when we look at published CPCs and
>> clinical problem solving exercises. Does this disparity imply that a
>> probabilistic approach is not suitable for diagnosis in actual practice?
>> This is certainly possible as diagnosis is performed in a given, individual
>> patient with the aim of determining a disease correctly in that particular
>> patient. And probability, as is well known has been employed most
>> successfully in practice in areas such as epidemiology and life insurance
>> business where the focus is on accuracy of prediction in a large group of
>> persons, not on prediction in a given individual person.
>>
>> If we look closely, we note that a strict probabilistic approach in which
>> a probability represents evidence may actually increase diagnostic errors
>> specially in patients with atypical presentations by encouraging the
>> cognitive bias of representativeness and inhibiting comprehensive
>> differential diagnosis (discussed in attached paper).
>>
>> I have put together my thoughts on this subject in the attached paper
>> ‘The role of probability in diagnosis’. Please review and comment on it.
>> Thanks.
>>
>>
>> Bimal
>>
>>
>>
>> Bimal P Jain MD
>>
>> Pumonary-Critical Care
>>
>> North shore Medical CENTER
>>
>> Lynn MA 01904
>>
>>
>>
>> ------------------------------
>>
>>
>> To unsubscribe from IMPROVEDX: click the following link:
>>
>> http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1
>>
>> or send email to: IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG
>>
>>
>>
>> Moderator:David Meyers, Board Member, Society for Improving Diagnosis in
>> Medicine
>>
>> To learn more about SIDM visit:
>> http://www.improvediagnosis.org/
>>
>>
>>
>> ------------------------------
>>
>>
>> To unsubscribe from IMPROVEDX: click the following link:
>>
>> http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1
>> or send email to: IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG
>>
>> Moderator:David Meyers, Board Member, Society for Improving Diagnosis in
>> Medicine
>>
>> To learn more about SIDM visit:
>> http://www.improvediagnosis.org/
>>
>
>
> ------------------------------
>
>
> To unsubscribe from IMPROVEDX: click the following link:
>
> http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1
> or send email to: IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG
>
> Moderator:David Meyers, Board Member, Society for Improving Diagnosis in
> Medicine
>
> To learn more about SIDM visit:
> http://www.improvediagnosis.org/
>






Moderator: David Meyers, Board Member, Society to Improve Diagnosis in Medicine


HTML Version:
URL: <../attachments/20151008/4328b012/attachment.html> ATTACHMENT:
Name: PhysicianCognitionLogo_email.png Type: image/png Size: 3718 bytes Desc: not available URL: <../attachments/20151008/4328b012/attachment.png>


More information about the Test mailing list