Sunday, August 29, 2010

Hospital Medicine: Out of Order



Physicans in Reverse Gear!


Here is some inside dope on the medical profession for patients to ponder.

We are all reading these days about improving the process of delivering medical care. This effort aims to raise the level of medical quality, and to minimize errors of omission and commission. This is why all surgeries and medical procedures begin with a ‘time out’, when there is a brief huddle confirming the identity of the patient and the intended operation. This is to prevent scenarios, such as:

“Mr. Patella, we replaced the wrong knee, but you would have needed a new one at some point. No need to get out of joint over this – the rehab is on us.”

Numerous medical specialties are now using checklists for medical procedures that include a series of steps. For example, if every heart bypass patient needs to proceed through 24 pre-operative steps, including laboratory studies, diagnostic tests, specialty consultations and an informed consent discussion, then a checklist is an effective tool to ensure compliance. Indeed, without a tracking mechanism, it is easy to understand how important steps can be omitted, with serious consequences. Chec-lists have been standard operating procedure in the airline industry, which have an excellent safety record.

Of course, there is a risk that physicians will become numb to all of these warnings and ‘time outs’. How carefully, for example, do we listen to flight attendants’ warnings and instructions prior to take-off?

This issue is relevant to how we physicians approach hospital patients. Here’s how we were taught to do it by our professors and mentors. See the patient first. Take the medical history personally, before you review the results of the CAT scan and other diagnostic tests. After taking the history directly from the patient, proceed with a methodical physical examination, which may provide important diagnostic clues. A medical condition that was considered probable after the history may be rendered unlikely after the examination. Additionally, an unexpected abnormality found on the physical examination, may lead the physician to pursue a different line of questioning that was not considered initially. After these two fundamental steps, the H & P, have been completed, the physician creates a differential diagnosis, a list of reasonable diagnostic considerations that can explain the patient’s condition. The doctor does not need to consider every diagnostic possibility, only those that are reasonable. For example, if you were to search diagnoses such as abdominal pain or fatigue on the internet, you could create lists that contain over a hundred entries. Sometimes, our patients bring us these lists convinced that their symptoms are explained by an obscure parasite, not present in this country, because the parasitic disease description matches the patient’s symptoms exactly.Obviously, physicians will expand the list of diagnostic possibilities, if the working list proves to be inadequate.

At this point, additional medical data are reviewed or tests are ordered to narrow the list, hopefully down to the correct diagnosis. Sometimes, the H & P is sufficient to make a reasonable diagnosis, and no further tests are required. For example, a patient who sees me with 6 weeks of heartburn that is consistently relieved with antacids doesn’t need to swallow my scope to make the diagnosis. I already know it, or I should.

This is how we were taught to see patients. Here's how it's done in the real world, particularly in the hospital. Every day, we physicians commit medical heresy, by seeing the patient last, after x-ray results, labs and specialist consultations are reviewed. I know this is true because I am one of these medical heretics. It is standard practice today to do what our medical school faculty beseeched us to avoid - seeing the x-rays before seeing the patient. Nowadays, when physicians enter patients’ rooms, we often already know that the CAT scan shows diverticulitis, or that there is a kidney stone, or that a cardiac stress test is abnormal.

What is wrong with this? Don’t physicians still get to the right answer, even if today’s diagnostic path is reversed?

I am hostile to this approach, which I practice, because it devalues the history and physical examination, which is the core of doctoring. When the physician greets a patient, and already believes that he knows the diagnosis, then the H & P became a diagnostic afterthought, a formality that must be performed more for documentation than for diagnosis. This means that the doctor will not take a broad and probing history, as he no longer feels the need to construct the classic differential diagnostic list. If the doctor meets a patient after viewing a CAT scan suspicious of appendicitis, he may take a cursory history and miss the correct diagnosis, which may not require surgery. It is in patients’ interest for physicians to think broadly at the outset, narrowing down the possibilities over time. It is harder to reverse this process. Once a patient is labeled with a diagnosis, it can be difficult to peel it off. This is why diagnostic labels should not be affixed prematurely. Seeing the patient out-of-order risks this outcome.

Internist to gastroenterologist: "Please see my patient, Mr. Calculus, in room 304 with vomiting. This is his 3rd gallbladder (GB) attack in 6 weeks.

This patient has now been labeled with GB disease.Will the gastroenterologist be open-minded enough to take a full history? Will he discover, for example that this patient ‘borrowed’ pain pills from a friend to relieve his back discomfor? Perhaps, his symptoms are a side-effect of these medicines. If so, then removing the gallbladder is the wrong move.

The out-of-order approach also has eroded physicians’ physical examination skills. If I already know from viewing a CAT scan that a patient has an enlarged spleen, then I will be biased when I palpate the abdomen at the bedside. Sure, I might feel the spleen, but will I examine the rest of the patient with necessary diligence? What if the spleen is not the critical finding? Physicians who examine hearts, lungs and abdomens – when they already know of abnormalities in advance – cannot be fully objective.

Finally, the out-of-order strategy reinforces to patients and the medical profession that testing and technology are more important than the initial human interaction with hands-on contact. This is not true, and yet this is increasingly the way of the medical world.

I maintain the medical history is the most valuable diagnostic took that exists, and it is at risk of being included on the endangered species list. Beyond its medical value, it is a foundation of the doctor-patient relationship, which is already under threat on so many fronts.
Here’s the checklist I wish I used without exception:

___ History
___ Physical
___ Other Stuff


Why do doctors like me who know that the H & P should come first, use the reverse gear istead?  The response to this question should be of great interest to the public, and I hope that this inside dope will appear below in the comments section.   OK, physicians, come clean.

15 comments:

A. Bailey said...

I always enjoy your essays, and I've enjoyed this as much as any. I'm old enough to have watched an elderly Rudolph Kampmeier perform an intricate and detailed physical examination, knowing that I would probably never be able to duplicate his technique. Legend has it that he would have a quarter baked inside a loaf of bread and he could locate it by percussion along, so fined tuned were his fingers.

Still, I doubt he could arrive at the proper diagnoses as quickly as we do with our technology. As good as he was, I doubt he could percuss out a colon polyp with high grade dysplasia.

Having said that, I think you are to some extent incorrect in the role we play as consultants. If I walk into a room and don't have every study that's been obtained in the last 12 months committed to memory, patients and especially their families think I'm ill-informed. They know the results of yesterday's CT scan; why on earth wouldn't I know it?

The paradigm is changing. We should not be trying to guess at what yesterday's CT scan might have shown. The real issue is this: what does it mean? Was the radiologist's interpretation possibly an over-call (that kind of thing is rampant in my community)? Do the findings actually relate to the patient's symptoms? Did the CT only reveal an incidental finding, while the true pathology remains undetected or unappreciated?

It calls for a different sort of discipline, and one we best all learn. ER doctors are not going to stop ordering CT scans on everyone with belly pain anytime soon.

The Medical Contrarian said...

I am not so sure the outcomes are better with the H&P first. We all need to review data on the patient first before the patient is sent to us (or we go to the patient). Obviously, at the most granular layer, you should not see a patient for evaluation of headache or a sprained ankle. That level of triage is a given. Once we buy into the idea that we can and should review patient information prior to our face to face encounter, what information is useful to review and why?

I generally review lots of information prior to seeing any new patient. I find I am much better prepared to fill in the holes in the history and get into the problem solving mode.
It appears that being prepared by reviewing all available data (particularly the history) prior to seeing the patient simply makes sense. The implication is that this approach risks some sort of diagnostic bias. All approaches will generates some sort of bias. I am not aware of any controlled studies which addresses which approach has better outcomes.

I also believe your commentary mixes two different types of tasks in medicine. The solution shop tasks are not as amenable to check list approaches as the value added tasks. A patient with abdominal pain cannot be diagnosed solely with a checklist approach. I agree that such an approach generates huge numbers of possibilities which really can only be sorted through using intuitive methods.

However, a patient with known disease (e.g. - gallstones) who is going for a defined procedure (cholecystectomy) will benefit from a checklist approach. It is foolhardy to rely on even very bright and highly paid professionals to consistently remember long lists of things that must be addressed in a reproducible fashion.

Elaine Schattner, M.D. said...

Hi Michael,
Interesting post -

I agree, it's a challenge to define the appropriate boundaries in medicine between when doctors should be thinking - critically and with an open mind - vs. behaving like robots.

GlassHospital said...

I misapprehended the title of the post, and thought you were going to write a screed against hospitalists.

I like the direction in which you went.

I wish I could share more dirt. To be honest, in academics, I find that even though my teaching involves exemplifying and preaching about the time honored tradition (i.e. history and physical exam first), I will often look at data (labs and scans) when I hear about a patient over the phone or at home when my resident team is on call.

I wonder if this is the reason we teach the trainees to do the time honored stuff first--so we can look at the "data stuff" and do our homework before rounds so that we look like the sharp attendings we're supposed to be.

Anyway, that's just two cents from someone still in academics.

-GH

Michael Kirsch, M.D. said...

All excellent comments! I too gather data in advance of entering the patient's room. Who has time to proceed with patient care in the classic manner of seeing patients first? Nevertheless, when I am scanning the radiology reports, etc. first, I feel I am diminishing the experience in some way. There's no question that technology has devalued the physician exam. Indeed, the CAT scan has become the abdominal exam. Similarly, who really knows how to ausultate these days?

DrJohnM said...

Good Stuff, doctor.

No better example of the history/PE trumping all other tests is in the evaluation of syncope. Give me a cogent story of an "episode," a physical exam and a 25$ ECG and I can save many thousands of dollars on MRI/EEG's and prolonged hospitalizations.

Keep up the good work. The blogosphere is rocking with useful information.

JMM

Jill--Unnecesarean said...

How funny! When I've described my experience in being transferred to the care of an OB in the 38th week of pregnancy due to suspected macrosomia after a late-term ultrasound, I've said that I walked in to someone sitting with her calendar open. Never having met me, the first thing she did was explain a few things and ask if I would like Sept. 1 or 2 for my c-section.

I found myself trying to tell her my "H" and even asked her to perform a "P". Now I know why that appointment felt strange. She went straight to "Other Stuff" without much more than asking my name.

Tom said...

I see your point. On the other hand, the patient often has already heard the test results before I meet them, and they get quite upset if I don't know what they are before I see them. Also, response to some test results should not be delayed while we go through a broad differential. Careful consideration of GERD or costochondritis is wasteful in the face of soaring troponin I.

Karyn said...

I am not a doctor, but I am an informed patient who is glad to have an explanation as to why I am always so frustrated when I go to the doctor.

I know my body better than anybody and am thankful when I meet a doctor who can recognize that and is willing to let me work alongside them to figure out the cause of symptoms and the best course of treatment.

Michael Kirsch, M.D. said...

Is it better for physicians to have the diagnosis in mind before meeting the patient, contrary to classical teaching? We gain in efficiency, but I maintain that there is a cost to this approach.

Nicholas Fogelson, MD said...

Michael - Great post. What you are describing is the idea of 'framing', where one person creates a context that another person will tend to follow. Jerome Groopman discused this in detail in "How Doctors Think", which was a great book.

Michael Kirsch, M.D. said...

Thanks, Nick. I think the 'backward' way sets us up to be duped. Say, a patient with a history of diveriticulitis is admitted with fever and diarrhea. The WBC is 18K and the CAT scan suggests left sided diverticulitis. The physician, who has not yet entered the patient's room, is understandably leaning toward a diagnosis. Perhaps, the patient truly has C. diff?

Anonymous said...

Also at the academic end, I found this a very interesting post. Mavis

B.RAD said...

Very interesting discussion. As a radiologist, I too feel that that sometimes the findings on a CT scan will lead the clinician down the wrong path-are those gallstones an incidentaloma or are they truly causing the patients abdominal pain? But in the real world, am I going to tell the EM doc to do the complete H&P, review all the labs, and THEN, order the scan? Of course this will add to the already long waiting times. I even get pressure to do the Abdominal CT scans without contrast (ie. no waiting for creatinine level, for oral contrast to traverse the gut), just to speed things up. The non-contrast study, of course, has it's limitations. This "reverse" workup is characteristic of our I.T. world.

Anonymous said...

Great comments. Many doctors are practicing what I call psychic assessment. In California, Southern California to be precise, I worked in a few hospitals where I found doctors were not assessing their patients. A physical exam? What's that? Doctors just take patient charts, sit down and chart their psychic assessments. I never once saw a doctor walking to a patient's room. I wish that I can be that good of a clinician performing a psychic assessment.

Add this