Missed and Erroneous Diagnoses Common in Primary Care Visits

Swerlick, Robert A rswerli at EMORY.EDU
Mon Dec 9 21:21:32 UTC 2013

Not to be a stickler but even the diagnosis of celiac disease is also not binary.  How does one define this? Some patients may have sub clinical gluten sensitivity for many years associated with no or vague mild symptoms. You might find low level anti-TG or gliadin antibodies in many patients with vague constitutional symptoms. Could these patients be simmering celiac disease? Perhaps. The transition from normal variant to disease state is not so clear. As tests for disease become more sensitive, the number diagnosed increases, but the actual treatment interventions may or may not change.  It is not just about the delay but also about whether earlier intervention makes any difference.

Obviously there are circumstances where diagnoses are flat out wrong and there were lost opportunities for earlier treatment and improved lives. In dermatology we see this in the form of a patient with undiagnosed scabies treated for eczema with systemic steroids for months to years when they could have been cured with topical permethrins is an example. This is similar to a patient with pernicious anemia in that a definitive and hopefully simple and safe intervention addresses a condition which has substantial morbidity and functional consequences.

How much diagnostic error as currently measured actually falls into this bucket is an open question.


From: Jason Maude [mailto:Jason.Maude at isabelhealthcare.com]
Sent: Monday, December 09, 2013 2:45 PM
To: Swerlick, Robert A; Society to Improve Diagnosis in Medicine
Subject: Re: [IMPROVEDX] Missed and Erroneous Diagnoses Common in Primary Care Visits

There are always lots of arguments as to why something won't work or various potential flaws in one particular case or another but there won't ever be a perfect answer so we need to find something that will get us most of the way there.

Why couldn't we take say a 1000 cases of a known diagnosis of celiac disease, for example (I am sure that there might be better ones to start with), and look at how long the diagnosis took from first presentation to definitive diagnosis?

With that we would know how long it took on average, the variation and then start to understand all the various factors that came into play leading to potential solutions. However until we know how long it does or should take how will we ever know whether the diagnosis was delayed other than just expert opinion?

Once this exercise is done for one disease we could then start to look at doing the same for other diseases and if we see the same pattern repeated across many diseases we could then extrapolate that across all diseases.


Jason Maude
Founder and CEO Isabel Healthcare
Tel: +44 1428 644886
Tel: +1 703 879 1890

From: <Swerlick>, Robert A <rswerli at emory.edu<mailto:rswerli at emory.edu>>
Date: Monday, 9 December 2013 18:23
To: Society to Improve Diagnosis in Medicine <IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>>, JASON MAUDE <jason.maude at isabelhealthcare.com<mailto:jason.maude at isabelhealthcare.com>>
Subject: RE: [IMPROVEDX] Missed and Erroneous Diagnoses Common in Primary Care Visits

from  my perspective it is not at all surprising that this data does not exist for a number of reasons. First, much of this data would need to be collected in outpatient settings where the precise collection of data is perhaps the lowest of all priorities. This is a major problem and will be addressed only when reliable data collection is integrated into normal workflows and taken out of the hands of physicians who attempt to do this while also trying to synthesize, make decisions, and write reports...simultaneous. No surprise we do none of this well.

Second, this perspective on diagnostic speed and accuracy appears to be based on a rather naive conception of what a diagnosis is. In my practice world, diagnoses are not exactly right or wrong, but more often more or less likely. Even after we have diagnosed and treated patients and they get better, there may still be substantial uncertainty regarding what actually happened. Although I do not have data to back this impression up, I suspect that a well-defined diagnosis is the exception rather than the rule.

There are so many contexts where diagnostic errors occur and we will never get anywhere in terms of improvement until we can come up with a better taxonomy. Some errors are caused by human failings, some by systems failings, some are due to limits of the diagnostic  tools we deploy,  and some by inherent limits on our knowledge. Some diagnoses must be made quickly and correctly the first time  while others will be prone to errors if they are made too fast. Some diagnoses were use are useless anyway and there are no consequences of getting them wrong.

I also sense that there are so many different priorities for diagnostic accuracy with inherently different priorities valued by clinicians vs. patients. I am struck by the patient stories which seem to highlight their frustrations with missed diagnoses over extended periods of time, during which they were afflicted with significant symptoms and functional compromise.  These stories tend to be in stark contrast to the stories of physicians who report the need to make decisions under substantial time constraints and limited or compromised data sets. Furthermore, in the absence of good information regarding how we fall short in the diagnostic realm, there is little to respond to other than those items which we are told can set us up for malpractice.

Yes, we should avoid diagnostic errors but for any given clinicians, where should we devote our efforts? How can we know we are getting better? I think the way forward must be to focus on better and automatic data collection tools which incorporate patient reported outcomes.

Bob Swerlick

From: Jason Maude [mailto:Jason.Maude at ISABELHEALTHCARE.COM]
Sent: Monday, December 09, 2013 4:09 AM
Subject: Re: [IMPROVEDX] Missed and Erroneous Diagnoses Common in Primary Care Visits

I think it would help to identify dx errors if researchers in this area pushed forward on looking at average time to diagnose for as many diseases as possible. If we could establish that on average it took, or should take,  x days to diagnose disease x from first presentation then, at least, we would would have some sort of bench mark that other cases could be compared to. Any delay of more than say 10 or 20% than the established norm would then be classified as an error. It seems odd that in 2013 we still don't know what the average to diagnosis for various diseases is.


Jason Maude
Founder and CEO Isabel Healthcare
Tel: +44 1428 644886
Tel: +1 703 879 1890

From: Ross Koppel <rkoppel at SAS.UPENN.EDU<mailto:rkoppel at SAS.UPENN.EDU>>
Reply-To: Society to Improve Diagnosis in Medicine <IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>>, Ross Koppel <rkoppel at SAS.UPENN.EDU<mailto:rkoppel at SAS.UPENN.EDU>>
Date: Sunday, 8 December 2013 19:01
Subject: Re: [IMPROVEDX] Missed and Erroneous Diagnoses Common in Primary Care Visits

Important to note the role of horrible decisions and actions by people/patients.
As a sociologist, I'd only add that there are a lot of societal factors creating some of that obesity, drink, cancer, lack of exercise, etc, etc. Many of those factors can be altered....but we don't

Ross Koppel, Ph.D. FACMI

Sociology Dept and Sch. of Medicine

University of Pennsylvania, Phila, PA 19104-6299

215 576 8221 C: 215 518 0134
On 12/8/2013 10:45 AM, Ted.E.Palen at KP.ORG<mailto:Ted.E.Palen at KP.ORG> wrote:
The death rate for climbers of K2 (the second highest peak in the world) is about 25%.  So although I understand your analogy it is not correct.  People take huge risks everyday. And they even take more risk with their health, although the results of their decision may not have consequences for many years. They smoke, drink to excess, become addicted to pain meds, eat poor, do not exercise.  Even though diagnostic errors and treatment errors may and do have terrible consequences, the behaviors people engage in have far more devastating consequences in regard to heart disease, cancer, obesity, diabetes, etc.

Ted E. Palen, PhD MD, MSPH | Physician Investigator | Institute for Health Research | Kaiser Permanente Colorado
Physician Manager for Clinical Reporting | Medical Cost Management| Colorado Permanente Medical Group
* 303-614-1215 | 7 303-614-1305 | *ted.e.palen at kp.org<mailto:sarah.madrid at kp.org>



This e-mail message (including any attachments) is for the sole use of
the intended recipient(s) and may contain confidential and privileged
information. If the reader of this message is not the intended
recipient, you are hereby notified that any dissemination, distribution
or copying of this message (including any attachments) is strictly

If you have received this message in error, please contact
the sender by reply e-mail message and destroy all copies of the
original message (including attachments).

HTML Version:
URL: <../attachments/20131209/9ceb516f/attachment.html>

More information about the Test mailing list