Wednesday, April 30, 2008

Hospitalist News launched

From the introductory article:

HOSPITALIST NEWS will cover breaking news in clinical trials, new therapies, and regulatory and payment trends that have a direct impact on hospitalist practice. Our experienced journalists will write concise, easy-to-read, and balanced articles that are fact-checked for accuracy. They will report from medical specialty meetings where new research is presented, monitor regulatory agencies, and provide the perspectives of thought leaders in patient care.

I’ll be linking from there often.

Hospitalist News.

Via Clinical Cases and Images blog.

CMS defines nine more hospital “errors”

Last year CMS, in an ill conceived policy change, rolled out a list of hospital acquired conditions which could no longer be coded to increase DRG reimbursement, defining these conditions as “never events.” The media egregiously spun the issue as “Medicare no longer paying for mistakes.” I pointed out how manifestly unfair this was in several posts last year.

Now they’ve added nine more to the list for consideration. Included in the new list of “medical mistakes” are iatrogenic pneumothorax, Legionella infection and delirium.

Well, I’m at a loss for words, so go read what Bob Wachter and DB have to say about it.

Poses vs Pitts on EBM

Peter J. Pitts, president of the Center for Medicine in the Public Interest, wrote a negative commentary in the Washington Times on evidence based medicine (EBM). Dr. Roy Poses, blogging at Health Care Renewal, offered a rebuttal. I’m not going to take sides in this debate, as good points were offered by both. Correctly noting that Mr. Pitts used a straw man argument (Pitts unfairly portrayed EBM as “one size fits all” medicine) and explaining what EBM really is (or is supposed to be) Dr. Poses then went on to base much of his own argument on an ad hominem attack against Mr. Pitts (it seems Pitts has ties to Big Pharma, who may not be happy with EBM’s objective evaluations of their new and expensive drugs).

“One size fits all medicine”, while not a core notion of EBM, is one of EBM’s popular distortions. We’ve seen it many times. Older and cheaper drug A performs as well in the latest systematic review as newer and more expensive drug B. Ergo, drug A is the drug of choice for all patients. This type of thinking may drive an agenda behind government funded research. Pitts cites the CATIE trial. An even better example (or at least one more familiar to me) might be ALLHAT. (DB of Med Rants has a great post on the apparent agenda behind ALLHAT here).

A larger issue is conflict of interest. Dr. Poses has written a great deal on conflicts of interest inherent in pharmaceutical industry funded research. But government funded research is conflicted too. The government has an interest n promoting cheaper drugs. Cheaper drugs are sometimes, but not always, better for patients.

Via Kevin M.D.

Tuesday, April 29, 2008

The medical record has been hijacked by non-clinicians

And the electronic medical record has facilitated the process. Dr. Wes weighs in on the electronic nursing note:

No doubt completing these charting requirements are simplified for today's nurses, but these nursing notes provide lines and lines of very little of substance for doctors to read, read exactly the same from patient to patient (and hence are ignored) and once something is found (like the social issues noted), no description of the issue is provided.Aspects of the new electronic medical record was not made for doctors or our patients, but clearly for quality assurance administrators. Thanks, folks.

Geriatric medicine declining as a specialty

Funny, that. I can remember, 20 years or so ago, when it was where hospital medicine is now---an up and coming specialty, the wave of the future. The reason for the decline? Bad reimbursement.

Via Kevin M.D.

The AAMC wants to ban the pens and pizza

----at all 129 U.S. med schools. Great. Now when are they gonna ban the woo?

Shameless hypocrisy.

Via Dr. Wes.

Retired Doc weighs in.

Friday, April 25, 2008

Managing insulin in the hospital

There’s a nice little review in Today’s Hospitalist on the subject. The diabetes Nazis have made the term “sliding scale” taboo. While basal-bolus regimens are more rational the “correction” component of short acting bolus regimens is nothing more than a sliding scale, as are popular insulin drip protocols. Moreover, basal bolus regimens are problematic in hospitalized patients:

While basal-bolus regimens, also known as physiologic insulin, are the preferred way to give insulin to patients in the hospital, these regimens bring their own set of challenges. Inpatients may be eating one minute, then being told to stop for a procedure, throwing up or being switched to an entirely different type of feeding.

The review presents practical tips to deal with these situations. It’s really a very helpful article but I had to chuckle at this:

At Loyola University Medical Center in Chicago, on the other hand, the protocol for patients with continuous tube feeds calls for one shot of slow-release glargine, plus a blood sugar check every six hours “and using correction factor dosing to lower an elevated glucose,” says Mary Ann Emanuele, MD, a professor of endocrinology and medicine.

The “correction factor” sounds a lot like a sliding scale although none dare call it that.

ECG electrode misplacement

It’s hard enough to read ECGs. Your job becomes even more challenging when the tech throws you a curve by misplacing the electrodes. This is the most systematic treatment of the topic I’ve seen.

Via Resident and Staff Physician.

Thursday, April 24, 2008

Hospitalists don the psychiatry hat

Psych beds are closing and psychiatrists are giving up hospital visits, leaving psychiatric inpatients underserved. Enter the hospitalist. Hospitalist groups, according to an article in Today’s Hospitalist, all too often become the default service for the care of these patients. But are hospitalists qualified?

When psychiatrists aren’t available, Dr. George points out, “some hospitalists are not comfortable prescribing the initial dose of certain psychiatric medications, like the newer antipsychotics.” While most hospitalists may be comfortable starting patients on antidepressants, “much further beyond that and their comfort level goes away.”

The problem surfaces in two ways. If the admitting hospital lacks psychiatric services the hospitalists often are called on to admit, bed and board the patient, attend to any acute medical needs and await placement in a suitable facility. If the hospital has psychiatrists on staff hospitalists are often called on to co-manage (there’s that buzz word again) those patients. Just as in the co-management of surgical patients, the arrangement creates problems if the lines of responsibility aren’t defined in stone.

What are the solutions? Some hospitalist groups have wisely demarcated the boundaries of their responsibility. Others are starting psychiatric hospitalist programs. The article discusses the ins and outs.

Near fatal asthma

The full text of a review appearing in Current Opinion in Pulmonary Medicine is accessible via Medscape.

Looking for a web based antibiotic guide?

There are usually several free pocket versions of the Sanford Guide lying around our place, left by the drug reps. I’ve found them difficult to navigate. The organization of content is suitable for electronic access but the book requires a lot of page turning. And if you can’t read the small font without a magnifying glass it’s just too cumbersome. I’d be willing to pay for electronic access but there’s no desktop version of the Sanford Guide.

An excellent alternative is the Johns Hopkins ABX Guide. Although dense in content it’s very easy to navigate due to extensive cross linking. You can browse by disease or drug, and can choose among brief summaries, tables and more lengthy monographs for each topic. There are extensive Q&A discussions and literature updates. The site is updated frequently---issues which have only surfaced in recent weeks are discussed. Handheld and desktop versions are included and the site is free after one time registration.

Wednesday, April 23, 2008

The debate over doctors and drug reps: a proposal

I’ve tried to be a voice for moderation in the debate. This proposal from Dr. James Alpert, editor of the American Journal of Medicine, is one of the better ones I’ve seen, but I don’t think it’s extreme enough for some.

Risk stratification in pulmonary embolism

Why is it important? Because there are now so many different management decisions. Is the patient a candidate for thrombolytic therapy? Does the patient need ICU? Is early discharge or out patient treatment an option?

Exciting research over the last few years has validated new tools for establishing risk. Right ventricular function assessment and cardiac biomarkers are useful. In February’s issue of Chest is a study showing high predictive value of the echocardiographic determination of RV to LV diameter ratio. An echocardiogram is often obtained in patients with PE. However, for a bigger bang for the buck, if the patient’s PE is diagnosed via CT, that study be used to obtain the RV to LV diameter ratio.

An editorial accompanying the Chest paper reviews the topic.

The electrocardiogram in hypothermia

Via the American Journal of Emergency Medicine.

Tuesday, April 22, 2008

Electrocardiographic findings in athletes

An electrocardiographic finding may indicate the effects of athletic training or underlying pathology which might contraindicate athletic participation. A review in the American Journal of Emergency Medicine offers tips in making the distinction.

Monday, April 21, 2008

The four hour antibiotic rule for pneumonia

Today’s Hospitalist has an update on antibiotic timing rules for pneumonia including the move by CMS and Joint Commission toward a six hour rule. Such a concession, in my view, will do little to alleviate the unintended consequences. The new IDSA guidelines have no timing rule, merely stating that antibiotics should be given as soon as possible in the ER.

The update mentions another problem with timing rules that has avoided the radar screen:

There is also the unproven but widely rumored suspicion that at least some emergency departments triage to favor potential pneumonia patients—even over individuals with other serious illnesses—so they will do well on publicly reported measures.


Sunday, April 20, 2008

Statin withdrawal in ischemic stroke

---was associated with worse outcomes in this study published in Neurology. Statin continuation was associated with lower rates of mortality and dependency. That’s better than TPA!

Distinguishing the causes of prominent electrocardiographic T waves

Prominent T waves may represent early (“hyperacute”) changes of STEMI, hyperkalemia or normal variant. The various patterns are reviewed in the American Journal of Emergency Medicine.

Friday, April 18, 2008

How to argue and why

The web, with its opportunities for interaction, lends itself to argument. How can we keep it constructive? Although there’s often a right and a wrong side to an argument seldom does anyone “win” by convincing the opponent to concede to a particular point of view. In most cases the best outcome is for both sides to gain a better mutual understanding, identify previously unrecognized areas of agreement, ultimately identifying the irreducible points where the parties may just have to “agree to disagree” and where further appeals to evidence and logic may or may not be productive.

Paul Graham, writing about disagreement in the cyber environment, has neatly schematized the analysis of argument by proposing a hierarchy of its various forms. The cheapest shot, level DH0 at the bottom of the list, is name-calling. As Graham points out, whether crude or articulate, name-calling is name-calling, and belongs at the bottom of the heap:

DH0. Name-calling.This is the lowest form of disagreement, and probably also the most common. We've all seen comments like this:

u r a #*@!!!!!!!!!!

But it's important to realize that more articulate name-calling has just as little weight. A comment like

The author is a self-important dilettante.

is really nothing more than a pretentious version of "u r a #*@."

Next up (DH1) is the Ad Hominem argument which, Graham notes, while occasionally useful, is still weak:

For example, if a senator wrote an article saying senators' salaries should be increased, one could respond:

Of course he would say that. He's a senator.

This wouldn't refute the author's argument, but it may at least be relevant to the case. It's still a very weak form of disagreement, though. If there's something wrong with the senator's argument, you should say what it is; and if there isn't, what difference does it make that he's a senator?

If this reminds you of the critics of drug detailing to physicians it’s because they resort to the ad hominem argument over and over again. (“Don’t listen to anything they say because they’re trying to sell something”).

And so the list goes, onward and upward to the top category (DH6) which is “Refuting the central point.” The scheme can help you refine your own arguments. Moreover, by putting them in easily identifiable categories it gives readers a tool to spot weaknesses in other people’s arguments:

The most obvious advantage of classifying the forms of disagreement is that it will help people to evaluate what they read. In particular, it will help them to see through intellectually dishonest arguments. An eloquent speaker or writer can give the impression of vanquishing an opponent merely by using forceful words. In fact that is probably the defining quality of a demagogue. By giving names to the different forms of disagreement, we give critical readers a pin for popping such balloons.

Raising the standard makes your argument not only more civil but also stronger. Cheap shots and personal attacks, while they make entertaining blogging, can weaken your effect. Astute readers may perceive, often correctly, that it’s the best you have to dish out.

The CreateDebate blog has illustrated the hierarchy as a pyramid to highlight the fact that the lower forms of argument are more common.

H/T to

Regret that tattoo you got years ago

---in a moment of youthful indiscretion? Dr. Tattoff can help.

H/T to

ER boarding and mortality

A report last June in Critical Care Medicine concluded that ER boarding of patients for greater than six hours awaiting ICU transfer is associated with increased mortality and length of ICU stay. I was tipped off to the article by this rant from Emergency Medicine News. And quite a rant it was:

Although reasonable solutions to this problem have been suggested (e.g., putting admitted patients on the halls of the wards), most administrators have chosen not to contaminate the rest of the hospital with these excess patients but rather hold them in the ED where they make the department malfunction substantially. Their reasoning: It may offend the medical staff, patients' families, and hospital floor staff. The philosophy is to sacrifice the ED and its patients for the sake of the rest of the hospital. Until recently, the perceived consequence of this decision was to reinforce the general beliefs of the community that EDs are associated with long waits and a litany of other

It sounds like more of the “us against them” finger-pointing I’ve commented on before. Here’s more:

We need aggressive medical staff leadership driving lengths of stay; we need administrators willing to close the hospital to elective surgery when the ED has no place to put its patients; we need options to efficiently discharge admitted patients; we need hospitalists committed to providing efficient, evidence-based care; we need ICUs open only to those qualified to admit patients to these units (rather than every Tom, Dick, and Mary on the medical staff). And, yes, some of these initiatives will upset the medical staff, but hospital and medical staff leadership need to jointly step up to the plate and take on the challenge.

ER crowding is a problem every hospital should be concerned with, and I agree with some of the suggestions above. But let’s get back to something more basic. From the introduction of the Critical Care Medicine paper (emphasis mine):

Emergency department (ED) “boarding” of critically ill patients (holding admitted patients pending ICU bed availability) is common and increasing in frequency in the United States, resulting in a prolonged ED length of stay (LOS).

Therein, perhaps, lies the problem. The patients were being boarded but were they being actively treated? Once the admitting doctor accepts the patient, that patient is considered “handed off.” There is a prevailing mindset that the ER is not a place for ongoing care once a disposition is reached. That mindset was expressed in the discussion section of the paper:

Whereas the ICU is a clinical environment that, by definition, enables close attention to the critically ill and allows for expeditious recognition of physiologic change and sudden deterioration, the ED under most circumstances is neither designed nor staffed to provide extended longitudinal care for the critically ill patient.

So, acknowledging that ER crowding is a problem with no quick fix in sight, perhaps, while we work on solutions, we should also look at what happens to ER patients during the delay. What kind of care are they getting? The study does not answer that question.

For some patients there’s evidence that definitive ongoing care in the ER for six hours before admission to ICU is associated with good outcomes. I’m referring, of course, to septic patients who are candidates for early goal directed therapy (EGDT). In the original protocol validated by Rivers, et al, patients were kept in the ER for EGDT. In fact, ICU personnel were not involved until the protocol was completed.

The solution is complex and multifaceted. Admitted patients who have no reason to remain in the ER for a specific intervention (e.g. EGDT) should be transferred as soon as possible. Inefficiencies in bed control should be sought and addressed. Ambulance diversion should be judiciously employed, Institute of Medicine recommendations notwithstanding.

This study raises another issue which hasn’t been addressed, and is ignored in the Emergency Medicine News piece: Given that ER crowding is going to be with us for some time, someone needs to address the quality of patient care during “boarding”, whether it takes place in the ER or other areas of the hospital.

Perioperative statin use

Evidence is mounting in favor of the perioperative use of statin drugs to reduce surgical risk. The ACC guidelines for perioperative evaluation and management of patients undergoing noncardiac surgery now give statins virtually equal status with beta blockers, recommending that patients already taking statins have them continued throughout the perioperative period.

I previously discussed perioperative statin use in a post from last September.

More recently this review of evidence and a summary of current recommendations appeared in Annals of Surgery.

Mortality of MRSA bacteremia in relation to MIC and vancomycin use

In MRSA bacteremia, when the vancomycin MIC exceeds 1 the mortality is increased in patients treated with vancomycin. The study was reported here in Clinical Infectious Diseases.

According to Medscape’s coverage of the study:

"These findings suggest that empirical vancomycin treatment when MRSA infection is suspected should be administered using a trough serum concentration of at least 20 mcg/mL as a target until a precise MIC is obtained, and it would be necessary to clarify whether new antistaphylococcal agents, such as linezolid, daptomycin, tigecycline or dalbavancin, could be superior to vancomycin when the strain has a vancomycin MIC > 1 mcg/mL," the authors conclude.

This study is from Spain. I’m not sure how it applies to the U.S., but I’ll be looking closely at the MICs of my patients with MRSA bacteremia.

Thursday, April 17, 2008


I thought I’d blogged enough for one day until I ran across this piece from Fox News. Many leaders in medicine are shaming us about global warming and related environmental issues. It’s even been suggested that psychiatrists subordinate the mental health of their patients to the health of the planet:

Therefore, instead of using psychiatric insight and techniques to reduce excessive anxiety, shame, and guilt for global warming these emotions will need to be increased in the unconcerned. This kind of 'help' runs counter to our usual goal of not making people feel worse!

Well, it’s working. The incidence of eco-anxiety appears to be rising. One sufferer is profiled in the Fox piece:

Sarah Edwards worries about the gasoline she burns, the paper towels she throws out, the litter on the beach, water pollution. She worries so much, it literally makes her sick.

The treatment? Hug a tree, take shorter showers, among other measures. Learn more at Sarah’s blog.

Harvard doctors speak out about downsides of electronic medical records

All too often they can be a substitute for thought, among other problems. White Coat Notes cited the NEJM article by Dr. Pamela Hartzband and Dr. Jerome Groopman, noting:

…computers make it too easy for doctors to lose focus on the patients before them. Residents and doctors can cut and paste one another's notes into the record, sacrificing the benefit of fresh eyes looking at a patient and distilling what is most relevant. Lab test results can flood the record with no selectivity on what matters for the current problem.

Computer template generated notes I’ve seen are often cluttered with boiler plate verbiage making it difficult to convey the patient’s story in meaningful form and nearly impossible to know what the doctor was thinking.

Via Kevin M.D.

Identifying heart failure from the get-go

This paper in the Journal of Hospital Medicine is more about the administrative diagnosis of heart failure than it is the clinical diagnosis. So is it even worth reading? Perhaps, but more for the clinical documentation specialists and core measure folks than for the clinicians.

How often have you seen this: The patient presents to the ER. Somebody casually mentions heart failure. Maybe the admitting doctor adds it to the differential diagnosis. The core measure team and clinical documentation specialists spring into action and clutter the chart with prompts and reminders. Come discharge time the patient, who turned out to have an exacerbation of COPD, not heart failure, is confused. She wants to know why a stranger came and told her to weigh herself every day, and what she’s supposed to do with the ream of paper on heart failure left on her bedside table. Or what about the patient whose correct diagnosis of heart failure was documented late in the admission resulting in a rush to comply with the core measures and the info packet being dropped in her lap as she’s wheeled out the door?

The paper presents a simple tool to help predict early on which patients will end up with a discharge diagnosis of heart failure. Show it to your core measure people. Maybe it’ll help.

Electrocardiographic manifestations of infectious diseases

Electrocardiographic changes are non-specific but may suggest a diagnosis or point to a specific complication. The topic is reviewed in a recent issue of Hospital Physician.

It’s yet another reason to do a “routine electrocardiogram” on really sick patients.

How much time should you spend washing your hands between patients?

Via a reader tip (my sister): Continually wash your hands while singing two choruses of Row, Row, Row Your Boat. Her source? Sixth grade science class.

Image source Wikipedka.

Wednesday, April 16, 2008

John R. Brinkley and the age of flimflam

There’s a new book out about John R. Brinkley. Harriet Hall has posted a review over at Science Based Medicine. In my post about Brinkley over a year ago I noted that we are much more tolerant of quackery now than in Brinkley’s day. In reading Harriet’s review (I haven’t read the book) I have to wonder if the author missed that point. Brinkley enjoyed success and wealth for a time, but much of this was gained in Mexico. He was virtually chased out of the U.S., losing not only his medical license but also his radio broadcasting license. When he constructed the first border blaster under a license from Mexico the U.S. tried once again to shut down his infomercials with passage of a law restricting the use of studio-to-transmitter links across international borders, known to this day as the Brinkley Act. While Brinkley’s surgical procedures would not be tolerated today his infomercials would thrive.

According to the review JAMA, decades ago, ran a regular quack busting feature. You won’t see that today in mainstream medical journals, which often promote quackery.

Tuesday, April 15, 2008

Old fashioned Grand Rounds

Not the medical blog carnival, but the traditional kind. More and more academic medical centers are archiving their presentations on the web. Here’s what I’ve found so far in Internal Medicine and related areas.

UTMB/JSC Aerospace Medicine Residency

University of Washington Television (Health and Medicine)

University of Nebraska

University of Florida

UC Davis

UA Birmngham



DAVE Project gastroenterology

FERNE presentations on neurological emergencies

UT Galveston

Wayne State University

Johns Hopkins Arthritis Center

University of Miami

NIH Clinical Center

UT Houston

UT Memphis

University of Arizona (until May 2007)

University of Arizona (After May 2007)

Ohio State

Drexel University

Heart Podcasts

POEM of the week podcast

Meharry Medical College


UT Memphis podcasts

The Canadian Medical Association Journal

---looks healthy, now over two years following the announcement of its collapse. It still has a lot of useful articles, which I link to often.

More negatives concerning erythropoiesis stimulating agents in hospitalized patients

According to a study presented at the Society of Critical Care Medicine 37th Critical Care Congress, the use of erythropoietin and darbepoetin in trauma patients was associated with a marked increase in venous thromboembolism.

Via Medscape.

Background here.

I don’t usually blog about local issues

---but I thought this little piece about hospitalist programs in my neck of the woods was worth a link.

Monday, April 14, 2008

Things overheard at the Society of Hospital Medicine Meeting

Here’s some more of the buzz from SHM 2008.

From Wachter’s World: How do you say no to relentless demand for hospitalist coverage? The scope of hospitalist coverage varies from one place to another. Hospitalists cannot be “house physicians” nor can they cover for every doc in town. My take on the issue is simple. Just say no. Your hospitalist group will be of no value to anybody if it burns itself out.

Also from Wachter’s World, related to the topic above: Recruiting, recruiting, recruiting! It appears most programs are looking to expand, driven by the constant need for more coverage. Great if you’re looking for a job, not so great if your program is short staffed. Despite the grandiose ideas expressed at SMH 2008 about hospitalists leading the health care quality revolution I suspect most programs are just struggling to meet day-to-day demands of patient care and call coverage. Every hospital has opportunities for improvement. For hospitalists to really take charge somebody has to carve out the time. That means lightening the patient load. Good luck convincing your administration.

From an attendee: In the wake of the impending CMS payment rules for hospital complications the new meme is POA (present on admission). Be on the lookout for the coming pandemic of decubs and urinary tract infections POA as the sensitivity for diagnosing these conditions goes way up. (Does one WBC per high power field really make a diagnosis of UTI?).

Again from Wachter’s World: Are you co-managing surgical patients yet? We’ve come a long way from the House of God where specialty wars and turfing (dumping the patient to a different specialty service) were the norm. Co-management is a new buzzword among hospitalist types. Collaboration between specialties is a wonderful idea, but one in need of better definition. Demarcation of responsibility remains important. All kinds of problems arise if it’s not clear who’s in charge of what. That’s my problem with the idea of co-management. What does it mean, exactly? Here’s my bias: Hospitalists can make sure patients’ statins and beta blockers get continued, help keep their electrolytes from getting screwed up, manage post operative cardiac problems and run ventilators. I’m not sure they should be fiddling with NG tubes, chest tubes or post operative pain management. However it’s done, demarcate the lines clearly.

The electrocardiogram in the patient with syncope

In addition to occasional demonstration of the culprit arrhythmia, the electrocardiogram may yield clues to underlying conditions such as acute coronary syndrome, Brugada syndrome, long QT syndrome and hypertrophic cardiomyopathy. The topic is reviewed in the American Journal of Emergency Medicine.

Saturday, April 12, 2008

ICU sedation

I’m not sure about the title of this article in Chest: Patient-Focused Sedation and Analgesia in the ICU. I thought everything we do is supposed to be patient focused. Anyway, it’s a helpful review which covers the ins and outs of various classes of agents and makes these key points:

Establish a specific rationale and specific goals at the start of treatment. (In other words, instead of “hurry up and snow the patient before the intubation meds wear off” take a few minutes to think).

Use a protocol.

Use daily sedation interruption.

Know your pharmacology.

Differentiation of narrow QRS tachycardias

An algorithm is presented in the American Journal of Emergency Medicine.

Clopidogrel rebound

When clopidogrel (Plavix) is stopped following a course of treatment for acute coronary syndrome, whether medically managed or with stent placement, cardiac events tend to cluster in the first 90 days after discontinuation according to a disturbing report in JAMA.

According to Medscape’s coverage of the study, the lead author commented that:

…there were two possibilities as to how to deal with it: keep the patient on clopidogrel for longer periods or taper the dose when stopping. "We need to study each of these possibilities to find out which one is best. But until we know for sure, physicians should discuss with each patient how they want to handle the situation after they have taken clopidogrel for a year after an ACS event. A patient who is doing well on clopidogrel, has no bleeding issues or other side effects, and can afford it may want to stay on the drug long term. If they want to come off the dug, then tapering the dose over a few weeks may be a good idea. Or perhaps the dose of aspirin could be doubled for a while. I can't recommend any of these things as we haven't got the data; they are just suggestions.”

Thursday, April 10, 2008

Big Pharma and the Society of Hospital Medicine meetings

The Society of Hospital Medicine regional and national meetings have grown rapidly in popularity due to the wide variety and high quality of their educational offerings. Although the success of these meetings is due in no small part to support from pharmaceutical companies, such support is controversial. Critics simplistically paint all industry support with the same broad brush, but there can be different degrees of involvement. At SHM 2008 several evening satellite symposia offered CME credit for a single topic presentation sponsored by a single company whose product related to the topic. Such activities are potentially problematic in contrast to the general sessions where industry support is broad based and insulated from meeting content.

Few would dispute that sales pitches and CME presentations whose topics directly relate to sponsors’ products influence doctors in favor of the products. It is also argued, but not supported by even a shred of evidence, that such promotions are harmful to patients. Although promotions may influence doctors away form best practice it is equally plausible that they influence doctors toward best practice. A case in point is that two of the seven satellite symposia were about DVT prophylaxis, a practice shown in study after study to be widely under utilized. The same was true for evidence based heart failure treatments such as beta blockers and ACE inhibitors back when those medicines were under utilized, on-patent and heavily promoted.

The critics, of course, claim to argue from evidence. Indeed, although there is no literature about patient outcomes much of the published material appears to show that the predominant influence on doctors is away form best practice. The problem with this large body of evidence, however, is that it’s biased. This bias manifests itself in the fact that the prescribing behaviors chosen for study are almost always ones known to be over utilized, e.g. antibiotic use and prescription of expensive antihypertensives. Nowhere in this vast repository of articles will you find surveys of doctors’ use of ACE inhibitors and beta blockers for heart failure, anticoagulation for atrial fibrillation or low molecular weight heparin for DVT prophylaxis.

But I digress. What about the general sessions? I believe adequate firewalls are in place to preserve the scientific rigor of the content. Next time you attend a national meeting of your specialty I challenge you to closely examine the general sessions and cite any content you think departs from best evidence to the potential harm of patients, and share your experience with me. And, if you don’t consider it beneath you, go to the exhibit hall and tell some of the reps you appreciate their support of the meeting.

Wednesday, April 09, 2008

Tuesday, April 08, 2008

Drinking the Kool-Aid at SHM 2008

(This is the first in a series of observations on the SHM 2008 national meeting. I enjoy SHM meetings and attend as often as I can. Unfortunately I missed this one. Reading the many blog posts and talking with colleagues who have returned from San Diego may be the next best thing. This, my inaugural post, somewhat critical, does not reflect my overall positive impression of this and other SHM meetings as will be evident from future posts).

Being the fastest growing specialty in medicine must be heady stuff. Notwithstanding the excellent “hard core” clinical content at Society of Hospital Medicine (SHM) meetings the organizational portions of the content, from where I sit, are looking more and more like a Wal-Mart Shareholders Meeting. (If you lived in the back yard of Wal-Mart corporate headquarters as I do you’d know that that gala event is one massive pep rally).

But take a look at some of the self congratulatory and grandiose ideas bandied about in the official blog of the 2008 sessions! One of the speakers, referring to the lofty (and, in the view of some, unrealistic) opinions espoused by IHI leader Donald Berwick, urged attendees to “’drink the Berwick kool-aid’ and insert ourselves in the center of the healthcare system transformation.” Reading some of those posts you’d think the hospitalist movement was what was going to “fix” health care. Enough already. We need to drink a little less Kool-Aid and engage in some critical thinking about what’s being proposed.

There was a refreshing voice of moderation. Bob Wachter, wise and all knowing sage of the hospitalist movement, said in his podcast that hospitalists should be thoughtful about future growth. In the past the agenda of the movement was to make hospitalists indispensable. Now, noted Wachter, we need to be careful that we don’t become too indispensable. We can’t do everything. The hospitalist movement is not going to single handedly “fix” anything. If we as hospitalist colleagues can just help each other ascend the learning curve of cost effective, science based hospital medicine in the care of individual patients we will do well!

Sunday, April 06, 2008

Prestige and money behind woo: The Institute of Medicine and The Bravewell Collaborative

Wallace Sampson’s recent Science-Based Medicine blog post about next year’s National Summit on Integrative Medicine and the Public Health, put on by the Institute of Medicine (IOM) and sponsored by the Bravewell Collaborative piqued my interest.

First a little background. In 2005 a committee of the revered IOM issued a report on complementary and alternative medicine in the U.S. I blogged about it at the time as being promotional of CAM’s irrational claims, citing Steve Barrett’s critical review of the report. Barrett noted:

Despite all the alleged experts involved in its preparation, the IOM report does not contain a single word of criticism against methods that are sufficiently irrational to be discarded now. Instead, it makes broad, sweeping generalizations and attempts to set an agenda for the widespread adoption of “CAM” research and teaching.

Barrett also discussed the report on his Quackwatch website in which he examined conflicts of interest inherent in the funding and heavy influence from the National Center for Complementary and Alternative Medicine (NCCAM).

The Bravewell Collaborative is a philanthropic organization which promotes questionable health claims in medical schools by funding the Consortium of Academic Health Centers for Integrative Medicine.

So what’s going on? Isn’t the IOM supposed give us a vision for quality health care? Whatever the vision is, it doesn’t appear to be a science based vision. Unfortunately the IOM’s promotions will lend an increasing appearance of credibility to unscientific health claims and encourage even more assimilation of woo into medical school curricula. As Sampson put it:

So “IOM,” in exchange for more $?millions as it did for the NCCAM committee, sells itself and its merit badge for “CAM”’s CV sash. Fair exchange in this capitalist system, yes?

How should we manage acute hypertension in stroke?

Current guidelines for ischemic stroke do not recommend emergency antihypertensive therapy until the systolic BP exceeds 220 or the diastolic exceeds 120 unless TPA is administered. The guidelines for hemorrhagic stroke allow a moderately more aggressive approach to blood pressure lowering depending on the clinician’s estimate or the actual measurement of intracranial pressure.

A new study suggests a change in this management strategy. Results from a small study (reported in Medscape) presented at the American Stroke Association's (ASA's) International Stroke Conference 2008 demonstrated that immediate antihypertensive treatment targeted to a systolic blood pressure of 145 to 155 mm Hg or a drop in systolic blood pressure of 15 mm resulted in improved stroke outcomes at 3 months.

The investigators acknowledged that the results of this small study should not change clinical practice and should be considered preliminary. Moreover, the study population was a mixture of patients with hemorrhagic stroke and ischemic stroke, limiting its applicability to either subgroup.

Cardiac memory phenomenon: a cause of non-ischemic T wave abnormality

Alteration in ventricular repolarization occurs during changes in ventricular activation as may occur with ventricular pacing, bundle branch block and various arrhythmias. This may result in T wave abnormalities which persist following cessation of pacing, resolution of bundle branch block or resolution of arrhythmia. Such changes may be confused with ischemia. The phenomenon, known as cardiac memory, is reviewed here in the American Journal of Emergency Medicine.

Don’t forget ECMO

Decades old studies suggested no improvement in outcomes for ARDS treated with extracorporeal membrane oxygenation (ECMO). Up until now the treatment has been viewed as a last resort. New evidence presented at the Society of Critical Care Medicine 37th Critical Care Congress (via Medscape) suggests it should be considered early in certain patients.

Most patients with ARDS die of causes other than hypoxemia, such as infection or the underlying disease that was a risk factor for ARDS. Those who appear at risk of dying of respiratory failure may be worthy of consideration for ECMO. The investigators used a Murray score of 3 or more as a criterion for patient selection.

The Murray score calculator is here. The study web site can be accessed here.

This evidence will undoubtedly cause a shift in thinking about ECMO. It remains to be seen how it should be incorporated into clinical practice, especially in hospitals lacking this capability. Should patients be shipped?

Friday, April 04, 2008

Base pneumonia treatment decisions on risk factors

A retrospective cohort study in Chest compared patients with community acquired pneumonia (CAP) admitted to an ICU with those admitted to a medical ward:

Conclusions: ICU patients present with more severe disease and more comorbidities. ICU patients stay longer in the hospital and have a much higher mortality rate when compared to ward patients.

While this was not particularly surprising or instructive the body of the paper contained important findings about the microbiology of CAP. True to prior experience, the etiology could be determined in only a minority of patients and among those with a microbiologic diagnosis Streptococcus pneumoniae was the most common pathogen. Other findings of note: Legionella antigen was not found in those patients tested and a surprisingly high number of Pseudomonas infections was found in both groups.

An accompanying editorial stressed that within the category of CAP are distinct risk groups that require different strategies. Some patients need a broader spectrum of antibiotics to include coverage for MRSA and Pseudomonas.

Current guidelines for both CAP and healthcare associated pneumonia (HCAP) recognize the need for risk stratification. Taken together the two guideline documents are potentially confusing: Whereas CAP and HCAP supposedly refer to distinct high and low risk categories with respect to the likelihood of difficult to treat pathogens, within both guideline sets are two or more such categories. Specifically, CAP guidelines stratify patients into those with high or low risk for pathogens such as Pseudomonas and MRSA whereas the HCAP guidelines stratify patients as high or low risk for “multiple drug resistant” (MDR) organisms. Antibiotic recommendations for the low risk categories in the two guideline documents are similar though not identical, as is the case for the antibiotic recommendations for the high risk categories of the respective documents.

I was lost when I first tried to sort through this. After preparing for this post, however, particularly after reading the paper and editorial in Chest, I began to appreciate the reasons for the multiple subtle distinctions. There are lessons here. Read the guidelines carefully with particular attention to the risk categories. In every encounter with a pneumonia patient ask yourself what risk category or categories apply.

Wednesday, April 02, 2008

More bad news about intensive glycemic control in critical illness

A large cohort study published in the journal Critical Care found no benefit, and a statistically insignificant trend toward increased mortality, associated with intensive glycemic control in ICU patients with a variety of critical illnesses. The glucose target in the most aggressively treated group was 80-110 mg/dl.

Can we close the gaps in care transitions?

In his Christmas Day post Bob Wachter had this to say about fumbled handoffs:

“…If transitional glitches were measured, and if botching them carried consequences, every hospital in the country would identify an accountable person in a nanosecond.”

Get ready to clear out the corner office of the C-Suite for the Chief Transitions Officer.

Although we’re not very good at washing our hands, we are terrific at washing our hands of patients who leave our medical radar screens.

That post contained a link to a wonderful resource which I had glossed over until recently: The Care Transitions Program. There you’ll find interesting vignettes, all sorts of implementation tools and a strong business case for a transition team.

Show it to your administrators. I did!

Hard learned lessons in EMR/CPOE implementation

What happened after the dismantling of CPOE at Cedars-Sinai in 2002 just three months after go live? They went back to paper orders and, it seems, just recently began preparations for reimplementation.

I had heard of the fabled EMR debacle and was curious about the details. So, today I dredged up the story from the Washington Post.

By January 2003, at a heated showdown with management, several hundred doctors demanded an end to the hospital's short-lived experiment.

If you’re involved in planning for adoption of CPOE at your hospital, read the article.

Emergency care of patients with implanted pacemakers and defibrillators: tips for the non-electrophysiologist

Given the growing number of patients with implanted cardiac support devices it is increasingly likely that hospitalists, emergency physicians and primary care physicians will face clinical decisions with these devices. A review in the American Journal of Emergency Medicine offers tips for evaluating patients and troubleshooting problems.

Tuesday, April 01, 2008

Checkout time 11 AM!

Busy hospitals around the country have put in place bed control initiatives to improve efficiency, reduce patient diversion, reduce ER wait times and reduce ER crowding. One such measure involves setting an early morning discharge time, analogous to check out time at the Hilton as Bob Wachter recently put it.

Pressure to discharge by a certain time of day is driven largely by administrators. Busy doctors struggle with it. Though conceptually flawed, difficult and loaded with unintended consequences (read Wachter’s post and comment thread) the idea has surface appeal. Clearing out the hospital in the morning opens up beds to alleviate the mid day crunch. That, in turn, reduces ER crowding and ambulance diversion. In some cases there are discharge planning advantages. Nursing homes and other long term facilities may not accept patients late in the day, and trying to arrange follow up appointments and obtain medical equipment presents difficulties when it’s 4 PM.

Vanderbilt University Medical Center (VUMC), where I attended medical school back in the days of the House of God, has become a model institution for systems improvements and efficient resource utilization. I’ve followed developments there through the years and read with interest about their discharge timing initiative, launched a few years ago:

VUMC faculty and staff are launching a patient discharge initiative to reduce midday patient access problems at Vanderbilt University Hospital. As recommended by the project team, the Clinical Enterprise Group (academic department chairs and senior administrators) endorsed two new requirements.
• Doctors enter “anticipate discharge” orders at least 24 hours ahead of discharge.
• Doctors enter discharge orders before 9 a.m. on the day of discharge.

On a typical busy morning those requirements could turn a doctor’s work flow upside down. According to project leader and professor of medicine Allen Kaiser:

He said writing of discharge orders before 9 a.m. will be a major change for faculty and house staff. As morning rounds get underway, doctors have always tended to visit the sickest patients first, so that for the less sick patient who is ready to go home, writing of discharge orders is typically delayed until later in the day. “It’s natural and very understandable for doctors to start the day attending to the sickest patients,” Kaiser said. “Switching to round first on the least sick patients is a gigantic culture change.”

And easier said than done. The notion of “rounds” (starting in one area and working your way from unit to unit) may be obsolete in this era of high acuity and much sicker patients. Hospitalists on busy medical services, constantly pulled in different directions, don’t often have the luxury of orderly rounds. Deteriorating patients in the ICU, rapid response calls and families demanding to see the hospitalist now (and threatening to go to administration) demand immediate attention. The wheel that squeaks the loudest at any given time gets the grease and it’s not optional. It’s too bad such events can’t be scheduled for the afternoon after discharges are complete.

When I read about the Vanderbilt initiative I thought it was onerous. Wachter’s post suggests that initiatives of this sort are popping up everywhere. Doctors at his institution (UCSF) are feeling the pressure and a similar program is underway at Stanford, where an administrator said:

Given these benefits, the hospital is launching an initiative to set our discharge time at or before 11 a.m. What can we do to help bring about this change? First, we need to make discharge decisions earlier in the day. This may require some process changes, such as conducting our attending rounds earlier.

Next, physicians need to plan ahead the night or the day before discharge

If it’s 3 PM and you realize your patient is stable for discharge why not discharge then rather than wait until the next morning? It may blow your time of discharge stats but it would help the hospital’s bottom line if the patient’s insurance plan (e.g. Medicare) doesn’t reimburse on a per diem basis.

Because setting an early discharge time may require keeping the patient an extra night hospitalists, also under pressure to cut the length of stay, find themselves under competing demands. Again, from Wachter’s post:

So when I am pressured to “improve” my time of discharge, I usually respond, “If you’d like, I can move the average discharge time up to 8 am. It’ll just be one day later than I had planned.” CFOs don’t like to hear that.Which brings me to my final plea: I believe it should be illegal to report Time of Discharge without also – in the same document – reporting adjusted average length of stay (or LOS against appropriate benchmarks). Time of discharge and ALOS are inextricably linked. The service that has a long length of stay AND a late discharge time might really have a problem. But the service with a short length of stay and a late discharge time is probably doing very good work, and harassing it over its TOD is annoying and counterproductive.

Finally, I chuckled at this from the Vanderbilt initiative:

Leutgens said plaques will be mounted on every hospital room door frame to inform patients of the VUH 10 a.m. discharge time.

Just like the Hilton.

Fasicular tachycardia

This is the rare exception to the rule that you never treat a wide complex tachycardia with verapamil. These patients are often hemodynamically stable. If you’re going to use verapamil, really know what you’re doing and don’t hesitate to get some help from a cardiologist or the EP folks.

Vanderbilt Medical Center critical care podcasts

These are from Vanderbilt’s burn ICU rounds but cover issues of general interest to critical care and hospital medicine. A new topic is posted every week or so. This resource is free of charge.

When your patient can’t afford low molecular weight heparin

Here’s a regimen for out patient (post hospital) treatment of VTE using subcutaneous unfractionated heparin without PTT monitoring.