Thursday, December 30, 2010

Physiology lectures from LA City College

These are word files and a few MP3s.

Partial DNR orders: always patient centered, often irrational, maybe harmful

In the era of patient centered care it's fashionable during DNR discussions to let people pick and choose from a menu of interventions, often resulting in “partial DNR” decisions. In some cases irrational combinations of interventions are chosen despite efforts to educate patients and family members on the rationale and expectations for various treatment modalities.

A recent paper in Critical Care Medicine reviewed the literature on partial DNR orders. Based on the limited published experience the authors suggested that the practice was “clinically and ethically problematic” and concluded:

Discouraging partial do-not-resuscitate(s) order may help promote more accurate and comprehensive advance care planning.

Osteoporosis treatment after hip fracture: how often done and by whom?

Not nearly often enough, and it depends on the specialty service according to this study.

Wednesday, December 29, 2010

The 2010 dumb awards for public policy speech

Bright people in high places occasionally say dumb things. In such a politically charged time as 2010 it made for great entertainment. Here are a few that made me laugh, mostly about health care reform. In no particular order.

Democratic Congressman Phil Hare on health care reform.

“I don't worry about the Constitution on this.”

And at 1:00 the Congressman confuses the Constitution with the Declaration of Independence.

Peter Orszag's New York Times column on October 3---

in which he implied (in a smooth way, but still strongly implied) that doctors don't work hard enough, doctors aren't used to having their performance measured, and hospitals shut down for the weekend.

A NEJM writer declared that if the Republican agenda on health care moved forward it would be a recipe for a failed republic.

Well, they won big in November. Better stock up on some survival provisions.

Nancy declared that we had to pass health care reform in order to find out what was in the bill.

Paul Krugman dropped the bomb about death panels.

This was such a bunch of double talk, it's hard to know what he really meant, but the plain point seems to be that we need to soothe the death panel alarmists now, then implement the panels later. Whatever he meant, it was dumb.

The interview with Bill Maher in which he said...

Well, he said a lot of dumb things. The whole clip is stupid. You'll just have to watch.

Tuesday, December 28, 2010

The end of life counseling debate---it's baaack!

Saturday's New York Times reports: Obama Returns to End-of-Life Plan That Caused Stir.

First a disclaimer. I am not against end of life counseling. In fact, as a hospitalist I often find myself wishing PCPs had spent more time discussing end of life issues with their patients.

On the other hand the political maneuvering and obfuscation by participants on both sides of the debate have long perplexed and concerned me. Boosters of the provision never adequately explained in plain language why it was necessary or exactly what it would add to what was already in place. Medicare payment for counseling was already embedded in the system of CPT codes. More than that, Medicare has for years allowed higher coding, exempt from the “key components” criteria for levels of service, if counseling was the principal reason (took more than 50% of the time) for the visit. That fact was never acknowledged in any of the debates or news reports I encountered. It raised the obvious question of why a new provision was necessary, and the question was never answered.

For me that was the key question and it was the one thing that made me just a little uneasy about the provision. Since we already had regulations in the CPT codes making it easy to get higher reimbursement for counseling, just what did the new provision add? Was it even better reimbursement? How much better? Did it elevate the topic of advance directives to some singular status above other types of counseling? After reading the provision in the original bill several times I was left with more questions than answers. No, there was no direct provision for death panels, nor was there any mention of pulling the plug on grandma. What was clear was that the mechanics would have to be worked out by policy makers with considerable discretionary authority.

This was supposed to be the administration of transparency. Transparency would have been helpful here. Unanswered questions, obvious questions, often lead to unhealthy speculation. Though not a death panel alarmist myself it's not hard for me to understand death panel alarmism in the public debate. It might have been averted had the proponents better articulated their position, and done so in plain and specific language.

The provision for end of life counseling was dropped from the original draft before final passage of the health care reform bill. So why is it resurfacing now? The New York Times article explains that the Obama administration has decided to implement the policy through what is known as the the regulation-writing process. It's a way to circumvent Congress---to enact the provision without legislative approval. The process isn't anything new. It's been used by Presidential administrations for decades. What's concerning is the effort to suppress debate. Again, from the Times article, referring to an e-mail from Democratic Representative Earl Blumenauer, one of the lead supporters of the provision (my bold):

Moreover, the e-mail said: “We would ask that you not broadcast this accomplishment out to any of your lists, even if they are ‘supporters’ — e-mails can too easily be forwarded.”
The e-mail continued: “Thus far, it seems that no press or blogs have discovered it, but we will be keeping a close watch and may be calling on you if we need a rapid, targeted response. The longer this goes unnoticed, the better our chances of keeping it.

That's what rubs people the wrong way. It's the transparency thing. This is an issue that needs clarity. Proponents didn't provide it the first time around. They owe that to the American public. An additional round of debate would offer them another chance to demonstrate clearly that this provision is nothing more than what they say it is: a means to facilitate informed treatment choices for patients.

Let me end with the perspective that, with or without the new provision, end of life counseling can (and always has) cut both ways. Depending on how honestly and skillfully the conversation is carried out (and it does take a great deal of skill) the outcome might be a better educated patient equipped to make an informed and rational choice, or, perhaps all too often, a poorly informed patient who is maneuvered into inappropriate exclusion of potentially effective care. The New York Times piece, quoting an academic palliative care specialist, gives these examples:

For example, Dr. Silveira said, she might ask a person with heart disease, “If you have another heart attack and your heart stops beating, would you want us to try to restart it?” A patient dying of emphysema might be asked, “Do you want to go on a breathing machine for the rest of your life?” And, she said, a patient with incurable cancer might be asked, “When the time comes, do you want us to use technology to try and delay your death?”

I'll give Dr. Silveira the benefit of the doubt and acknowledge that the Times may have misrepresented her statements, but these are examples of misinformation, inappropriately leading the patient away from beneficial care. If this is how an academic palliative care specialist, arguably the best among the experts in such counseling, informs patients about their options, how is it done in the real world? This is where the law of unintended consequences kicks in.

So what about the heart attack followed by heart stoppage (usually VF)? It would be inappropriate to expect most patients to give an informed answer to that question without explaining the very high success rate for defibrillation when VF is witnessed in a hospitalized patient experiencing MI. As for emphysema, mechanical ventilation does not mean going “on a breathing machine for the rest of your life.” A course of mechanical ventilation for COPD and respiratory failure is a valid option for many patients, is often of short duration and carries a reasonable prognosis.

What did we learn about patient safety in 2010?

St. Louis City Hospital

Popular belief holds that due to a “movement” that was launched about a decade ago, patients in hospitals are safer. Although patients are better off year by year due to incremental improvements in medical knowledge and technology there has been little published evidence up to now regarding any impact of the patient safety movement. I blogged last year, following a negative report by the Consumers Union, that the movement was a failure.

Then last month a study was published in NEJM showing no improvements in patient safety over a several year period during the decade of the movement. Following this disappointing report there has been much speculation as to the reasons for the failure. While some ideas, notably central line and perioperative check lists lived up to their promise, others such as CPOE and medication reconciliation, while elegantly simple in concept, failed because they were full of unintended consequences and inscrutably hard to implement in real practice. Other measures such as hand washing, due to their implementation as performance measures, may have had little more than cosmetic impact as I explained here.

I have also maintained for several years that the Institute of Medicine (IOM) report on patient safety, credited by many with launching the movement, actually harmed the cause. Patient safety experts, citing the need for transparency to facilitate discussion and analysis of adverse patient events, were calling for a culture of reduced blame. The IOM report, with its sensationalistic claim that up to 98,000 patients died in the US every year from medical errors, had the opposite effect.

Because the IOM is a subsidiary of the National Academy of Sciences most observers viewed the report uncritically, assuming that it was a scholarly work, despite heavy criticism in the scientific literature and a repudiation of the claim by the lead author of the Harvard Medical Practices study upon which the IOM conclusion was largely derived. In fact, the IOM report was anything but scholarly. It was a massive publicity stunt. Bob Wachter, one of the report's leading enthusiasts, even characterized it as a piece of masterful spin. Its claim of 98,000 deaths by medical error was no more scholarly or fact-based than Sarah Palin's claim about death panels. Despite its flawed nature the popular press gave it a life of its own. Far reaching effects on public perceptions and policy soon followed. I recently examined the public policy consequences of the report and the manner in which it helped produce a culture of provider blame which undermined the cause of patient safety in a guest post at Kevin MD.

Hospitalists at the close of 2010: how are we doing and where are we headed?

St. Louis City Hospital

Hospitalists and outcomes
Boosters of the hospitalist movement have been saying for years that the hospitalist model of care resulted in cost savings. Until this year that claim was not evidence based. Although several studies and systematic reviews had suggested improved efficiency attributable to hospitalist care one of the largest and methodologically best studies on the model showed no difference in efficiency attributable to hospitalist care. But after making a brief splash in the blogs in 2005 it was conveniently tossed into the memory hole. A large new study out this year appears to have tipped the balance of evidence in favor of the hospitalist model by showing reduced length of stay, with an increase in the magnitude of the effect over time through the study period.

There were two items of interest in this year's literature in the related field of comanagement. When hospitalists served as attending physicians for patients admitted with GI bleeds (and thus comanaged the patients alongside the gastroenterologists) costs were higher and there was a borderline statistically significantly higher rate of readmissions in a study I discussed here. Another study just out this month looked at the experience with comanagement of neurosurgical patients at UCSF, which I noted here. According to the results of that study, although the perceived value of comanagement was high, objectively the service appeared to be value neutral.

Current recession and health care reform
Hospitalist salaries have continued to climb, with no sign of moderation, right through the recession. Hospitalists are being paid what the market demands and this is driven by a shortage in the work force. This was documented in the first ever combined MGMA-SHM survey, the results of which I referenced here. The trend will eventually moderate, but no one knows when or at what level. It will likely continue as long as hospitalists remain in short supply. An early influx of hospitalists seeking jobs came from traditional general internists making a career move. That pipeline is likely to run dry soon. Among IM trainees leaving residency, although many take temporary hospitalist positions before landing a subspecialty fellowship or making other definitive plans, the percentage of graduates seeking a hospitalist career is relatively small. Meanwhile, according to 2010 American Hospital Association survey data, 43% of hospitals were making efforts to increase the number of employed hospitalists.

It's anybody's guess what the impact of health care reform will be. It is in a shambles now and we don't know what it will look like in its final form. However, if currently developing economic pressures continue hospitals will need hospitalist programs in order to survive. Although the number is likely much smaller now, according to survey data from almost two years ago nearly half of hospitals did not have hospitalists. So until this growing niche is saturated hospitalist salaries are likely to continue to climb. On the other hand hospitalist jobs could be negatively impacted if large numbers of hospitals are shuttered by health care reform. While that's an outcome that may seem desirable to some policy makers it's not a likely scenario in the near term.

The evolving (or devolving) hospitalist job description
As I said in a post earlier this year, My how we've changed. Cultural and economic shifts have altered the hospitalist's job description over the past decade, and it's not for the better:

Hospitalist skill set, year 2000

Management of CAP and HCAP
Management of sepsis
Management of venous thromboembolism
Management of DKA, adrenal crisis and other endocrine-metabolic emergencies
Management of COPD, asthma and ARDS, including ventilator management
Management of renal, fluid and electrolyte problems
Management of acute decompensated heart failure
Management of toxicologic emergencies
Providing consultation for complex and difficult diagnostic problems
Providing consultation for medical complications in surgical, subspecialty and general medical patients

Hospitalist skill set, year 2010

Tweaking Press Ganey
Tweaking clinical documentation (DRGs)
Creative coding
Bed control
Admitology, roundology and dischargology for surgical and subspecialty patients
Cosmetic charting (performance measures)
Secretarial (CPOE)

What's new in CPR and emergency cardiac care in 2010 and what does it mean for hospital systems of care?

St. Louis City Hospital

New ACLS guidelines came out last fall. I posted a brief discussion with links to the original document here. For the management of cardiac arrest the guideline changes fell short and remain years behind the best resuscitation science. Welcome changes, though, came in the recommendations for post resuscitation care and stroke care.

In the past, new ACLS guidelines merely meant more material for providers to learn. Not so in 2010. Now, for the first time, ACLS guidelines mean major system changes for hospitals.

A post resuscitation bundle to include applied hypothermia and urgent PCI in many patients is now a class I recommendation:

A comprehensive, structured, multidisciplinary system of care should be implemented in a consistent manner for the treatment of post–cardiac arrest patients (Class I, LOE B). Programs should include as part of structured interventions therapeutic hypothermia; optimization of hemodynamics and gas exchange; immediate coronary reperfusion when indicated for restoration of coronary blood flow with percutaneous coronary intervention (PCI); glycemic control; and neurological diagnosis, management, and prognostication.

Neurological prognostication following return of spontaneous circulation (ROSC) has been turned upside down in the hypothermia era. Traditional clinical rules apply for patients not treated with hypothermia. For those patients treated with hypothermia all bets are off, and here's the new class I recommendation:

There is a paucity of data about the utility of physical examination, EEG, and evoked potentials in patients who have been treated with induced hypothermia. Physical examination (motor response, pupillary light and corneal reflexes), EEG, SSEP, and imaging studies are less reliable for predicting poor outcome in patients treated with hypothermia. Durations of observation greater than 72 hours after ROSC should be considered before predicting poor outcome in patients treated with hypothermia (Class I, Level C).

Part 9 of the new guidelines is a useful reference on many aspects of post resuscitation management.

The section on stroke care (part 11 of the guidelines) also calls for system changes. Transport of stroke patients to a designated center and admission to a stroke unit are class I recommendations.

Monday, December 27, 2010

Respiratory failure and mechanical ventilation---anything new and important in 2010?

St. Louis City Hospital

Although there were no landmark trials reported in 2010 several important trends emerged in the literature.

Severity assessment in ARDS
Like several other conditions in hospital medicine (e.g. C diff disease and pancreatitis) severity assessment has treatment implications in ARDS and was a recurring theme in 2010. The major tool for severity assessment is the lung injury score (calculator here) although other methods exist. This year there were three papers dealing with special therapeutic modalities to be considered in patients with ARDS who meet severity criteria. One looked at an old modality considered, in recent years, to be outmoded: the use of paralytic agents added to sedation for severe ARDS. This could represent a swinging back of the pendulum and is likely to remain controversial for some time, at least until further systematic study is done. I discussed the NEJM paper here.

Another paper this year, discussed here, mentioned a meta-analysis suggesting that high levels of PEEP (above those recommended in the ARDSnet PEEP scale) may be beneficial in patients classified as having severe disease.

High PEEP can be considered a rescue therapy. Rescue therapies have not been validated across the board for most patients with ARDS. A recent paper (discussed here) looked at several rescue therapies including high PEEP, prone positioning, high frequency oscillation and extracorporeal life support. These are therapies to be considered in severe disease. Although considered last resort measures which may offer the patient's only chance at survival they may warrant consideration and preparation early in the course of some patients whose presentations meet severity criteria.

Low tidal volume ventilation---not just for ARDS anymore?
There is accumulating evidence that, other things being equal, concerning delivered tidal volumes in mechanical ventilation, the lower the better. Early this year I reviewed the literature on this subject and noted the first RTC to suggest that low tidal volume ventilation was beneficial in patients who did not have ARDS/ALI at presentation, because it reduced the incidence of ARDS/ALI developing during MV.

Special problems in mechanical ventilation of patients with heart disease
Known or occult heart disease presents unique problems during mechanical ventilation and can be a cause of “failure to wean” (FTW). Although this has been known for some time it is underappreciated in day to day critical care practice and seemed to get special attention this year. Awareness of the problem is being raised. Briefly, the pathophysiology of cardiac disease as a cause of FTW relates to the multiple hemodynamic benefits of mechanical ventilation which, when withdrawn, may result in acute myocardial ischemia and cardiac decompensation. Due to the special circumstances of mechanical ventilation cardiac ischemia and decompensation may not be easily recognized. Judicious use of cardiac markers and even serial echocardiograms during spontaneous breathing trials may aid the clinical assessment. I discussed papers dealing with this problem here, here and here.

How can we leverage adherence to best practice in the hospital? New findings in 2010

St. Louis City Hospital

Poor uptake of evidence in clinical practice has been a stubborn problem. Many solutions have been offered, with disappointing results. Up until this year the impact has been confined mainly to process measures and has been modest at best. But a paper published early this year reported on a process improvement initiative which more than doubled the rate of adherence over two years. More than that, it resulted in an improvement in mortality and it came from, of all places, industry. It was, of course, the Surviving Sepsis Campaign.

I've hammered at this issue before and pointed out that the Surviving Sepsis Campaign, described by NEJM Perspective piece writers as little more than marketing disguised as evidence based medicine, is one of the most maligned promotional campaigns in all of industry. The lesson here is that industry collaboration, though not a politically correct option these days, may provide opportunities to leverage better adherence to evidence.

2010 was a good year for biomarkers in hospital medicine

St. Louis City Hospital

Are you using biomarkers for all they're worth in your hospitalized patients? Here's a summary of new developments this year.

Procalcitonin has been validated for several roles in guiding the use of antimicrobial therapy, and stands poised to bring the concept of antibiotic stewardship to a new level. The major barrier is its lack of availability in real time in many hospitals. It has recently been approved for use in the US. I linked to a paper reviewing today's use of procalcitonin, here.

BNP and pro-BNP
The precise role of natriuretic peptides in guiding heart failure management remains a subject of controversy.   But the most interesting developments this year were in the area of expanded uses of natriuretic peptide measurement outside the management of heart failure. Perhaps the best known of these is risk assessment in patients with pulmonary embolism, discussed in several papers this year and in previous years.

Two new papers were of special interest this year. One looked at the kinetics of pro-BNP in acute onset atrial fibrillation. In cases where the time of onset of a-fib is unclear measurement of pro-BNP may help. Another paper, which I discussed here, reported preliminary evidence that the use of pro-BNP combined with a clinical risk assessment tool in patients presenting with chest pain produced outcomes equivalent to the conventional strategy of “rule out” followed by stress testing.

The role of lactate measurement to assess septic patients' candidacy for early goal directed therapy was validated several years ago. This year produced new information on expanded indications for lactate, particularly the measurement of serial levels in critically ill patients. Two papers addressed this issue, which I linked and summarized here.

Hyponatremia in 2010---new and noteworthy

St. Louis City Hospital

When patients are hospitalized with hyponatremia, either as the principal problem or an incidental finding, the hospitalist is immediately faced with several questions. Does the patient have acute hyponatremic encephalopathy or not? Should the serum sodium be raised? How much and how fast? What is the best strategy to avoid osmotic myelinolysis?

Several important papers were published this year which may help refine the approach to the evaluation and management of hyponatremia. For acute hyponatremic encephalopathy bolus therapy with 100 ml at a time of 3% saline is gaining popularity and has sound rationale (see here).

The past year saw increased published experience with the use of desmopressin to blunt the rise in serum sodium and decrease the risk of myelinolysis, either reactively in situations of threatened overcorrection or proactively as part of the initial treatment, at least in cases of very severe, e.g. “double digit” hyponatremia.

One paper this year emphasized the special hazard of hyponatremia accompanied by hypokalemia and reminded us that correction of hyponatremia, and possibly even overcorrection, may result merely from efforts at potassium repletion.

Finally, two papers addressed the etiologic diagnosis of hyponatremia, pointing out that clinical volume assessment is difficult and discussing some fine points in the evaluation of serum and urine chemistries.

Sunday, December 26, 2010

What were the most important issues in hospital medicine in 2010?

St. Louis City Hospital

It's popular right about now to look at the year in review. For the field of hospital medicine I have traditionally posted a top ten list. Although I'm dropping the top ten format this year (it tends to be artificial) I'm preparing some posts for the next few days on what I think the most important developments were in 2010. Among those are updates in safety, biomarkers, mechanical ventilation, cardiac function in critical illness and resuscitation. Stay tuned and look for the hospital sketch above.   

Friday, December 24, 2010

What's going on with the National Health Service?

The health system Don Berwick is in love with is anything but patient centered, and on the verge of a total melt down from the sound of this article from the Daily Telegraph.


Bob Wachter, meanwhile, has a different take (essentially that the Brits accept it, more or less, due to their inherent trust in big government and the elites, in a way that could never be so in the US).

Hospitalist comanagement of neurosurgical patients: loved by all and value neutral

One of my ortho colleagues gently taunts: “You're supposed to be the smart doctor. I'm just a dumb bone cruncher.” But I'm just as dumb about all that bone crunching surgical stuff as he claims to be about internal medicine. And so betwixt us both the bases get covered. Or so we hope.

Even in the days of old one of my close surgical colleagues and I were doing comanagement, long before anyone coined the term.  We didn't have cardiologists in those days, so I selected patients for pacemaker implantation and he performed the implants.  He would say "You take care of the tachyarrhythmias, I'll take care of the bradyarrhythmias."

Those are not very good definitions of hospitalist comanagement but it's about as precise as it gets. Which is why it's very difficult to draw conclusions from research evidence.

A recent issue of the Archives of Internal Medicine has published findings from the comanagement experience on the neurosurgical service at UCSF:

Results  During the study period, 7596 patients were admitted to the neurosurgery service: 4203 (55.3%) before July 1, 2007, and 3393 (44.7%) after comanagement began. Of those admitted during the postimplementation period, 988 (29.1%) were comanaged. After implementation of comanagement, no differences were found in patient mortality rate, readmission, or length of stay. No consistent improvements were seen in patient satisfaction, but strong perceived improvements occurred in care quality reported by nurses and nonnurse health care professionals. In addition, we observed a reduction in hospital costs of $1439 per admission.

But the more I dig into the body of the paper the more confusing it gets. There were lower costs in the post-comanagement as opposed to the pre-comanagement period (of only borderline statistical significance) but when comanaged patients were compared directly with non-comanaged patients (raw data not displayed in the paper) the comanaged patients had higher costs. To confuse things further, the authors say these were adjusted outcomes but in the same sentence imply that the difference was due to lower complexity in the non-comanaged patients. (OK, I'm very tired after a difficult few days of comanaging patients myself, so I may have missed something here).

And what about value? According to Skeptical Scalpel, quoting from an interview of one of the authors, the reimbursement to the comanagement service by the hospital was substantial.

Bob Wachter, one of the authors of the paper, gives an insider's view of the comanagement effort here.

Thursday, December 23, 2010

Blackberry not working? There's help.


Clostridium difficile update

Scanning EM of C diff

This review (free full text) was recently published in Southern Medical Journal.

Many C diff reviews have been published in the last few years. This one emphasized several points not covered in previous reviews:

Community acquired C diff is a newly emerging problem. Many patients have no history of recent prior antibiotic usage and lack other traditional risk factors.

C diff can occur in the setting of inflammatory bowel disease. IBD, in fact, is listed as a risk factor for C diff.

Mild disease can evolve into severe disease during treatment.

Leukocytosis may precede GI symptoms, so unexplained leukocytosis developing in a hospitalized patient may be a red flag.

C diff may present as acute abdominal symptoms without diarrhea.

Toxin assays have a sensitivity of only around 70%. Two negative toxin assays in a row, however, are associated with a high negative predictive value.

First recurrences are treated with the regimen to which the patient initially responded unless the recurrence is with severe disease in an patient previously treated with mild disease.

Subsequent recurrences are more problematic. Tapering regimens of vancomycin, pulsed dosing of vancomycin, vancomycin combined with rifampin, or vancomycin followed by rifaximin have been described.

Because alcohol based hand sanitizing agents are not sporicidal, hand washing with soap and water is recommended when caring for such patients.

Review of cardiac amyloidosis

Free full text from JACC:

Although no single noninvasive test or abnormality is pathognomonic of cardiac amyloid, case-control studies indicate that echocardiographic evidence of left ventricular wall thickening, biatrial enlargement, and increased echogenicity in conjunction with reduced electrocardiographic voltages is strongly suggestive of cardiac amyloidosis.

Via The Bottom Line.

Tuesday, December 21, 2010

Another SVT mimic

I recently mentioned 1:2 AV conduction. Here's another one: interpolated junctional extrasystoles. The authors considered 1:2 conduction in each case but it turned out to be junctional firing interpolated between sinus beats, leaving the sinus mechanism undisturbed, behaving as a parasystolic focus. An abrupt, exact doubling of the ventricular rate provides a clue.

Pylephlebitis as an unrecognized cause of septic shock

Case report in the Southern Medical Journal and brief discussion at The Bottom Line.

Why do hospitalists and IM house staff order PPIs?

Often for the wrong reasons in this study:

Sixty-nine percent of physicians reported prescribing SUP to greater than or equal to 25% of patients. In multivariable analyses, the following factors were associated with higher level of prescribing (greater than or equal to 25%) of SUP: fear of gastrointestinal bleeding (OR = 2.7, 95% CI 1.07, 7.28) and of the legal repercussions of not prescribing SUP (OR = 3.02, 95% CI 1.07, 8.56), whereas knowledge of SUP indications (OR = 0.39, 95% CI 0.20, 0.74) and concern about side effects (OR = 0.24, 95% CI 0.09, 0.61) were associated with low prescribing behavior. Level of training was not associated with prescribing rate. Less than half of respondents were able to identify a single side effect of proton pump inhibitor therapy.

Monday, December 20, 2010

ER-hopping drives up costs

I'm getting tired of going to Regional. Let's try St. Joe's tonight!

For many patients hospitals are like restaurants.

A new study in the Archives of Internal Medicine showed that such hospital-hopping is common and it drives up costs. And there's reason to believe it's unsafe. You can take for granted that medication errors will increase, and in this study patients who went to multiple ERs were more likely to be hospitalized overnight.

Occasionally a trip to a different hospital is unavoidable but in general, if you want to do your part for the medical commons and make your hospital visit a safer experience don't hospital-hop!

Via Grunt Doc.


Southern Medical Journal has published a two part series on adverse drug reactions, available as free full text here and here.

Acute RV failure understudied and underappreciated

Here's a review from JACC. Full text available via Medscape here. We usually think of pulmonary embolism and RV infarct but there are many other causes. Management is not as straightforward as you might think, and goals of therapy may conflict with those of coexisting disease processes in the critically ill patient.

In which Thomas Stossel smacks down the arguments against physician-industry collaboration

Via Med Policy.

Update on short QT syndrome

This condition has gotten a lot of attention in the last few years, but as far as we know it's rare.

Friday, December 17, 2010

Interview with David Sackett, one of the EBM movement's founders

Although EBM (at least in its popular application) cannot evaluate the unscientific claims of CAM (you need SBM to do that) it is useful for evaluation of most of the clinical questions that arise in mainstream practice. And I wonder what David Sackett would say if asked about EBM's blind spot. Right off the bat in the interview below he swats away some of the concerns of EBM's detractors by saying that first and foremost it's about the patient and clinical expertise. It's these elements that must be combined with the “best external evidence.”

His review of EBM's history is interesting. The movement had its antecedents in the 1960s with the birth of the discipline of clinical epidemiology at McMaster University.

“The core of her appeal is her deep love for freedom”

Anne C. Heller on Ayn Rand.

Via Instapundit.

What do docs in the rank and file think of health care reform?

Here are some survey results released by the Physicians' Foundation:

• The majority of physicians (60%) said health reform will compel them to close or significantly restrict their practices to certain categories of patients. Of these, 93% said they will be forced to close or significantly restrict their practices to Medicaid patients, while 87% said they would be forced to close or significantly restrict their practices to Medicare patients.
• 40% of physicians said they would drop out of patient care in the next one to three years, either by retiring, seeking a non-clinical job within healthcare, or by seeking a non-healthcare related job.
• The majority of physicians (59%) said health reform will cause them to spend less time with patients.
• While over half of physicians said health reform will cause patient volumes in their practices to increase, 69% said they no longer have the time or resources to see additional patients in their practices while still maintaining quality of care.

Via Drugwonks.

Acupuncture: hard core woo

As long as we're on the subject, read Steve Novella's recent take down at Science Based Medicine.

Thursday, December 16, 2010

EBM's blind spot

The Cochrane Collaboration, arguably the exemplar of EBM, is a good place to illustrate EBM's blind spot, a phrase used by Kimball Atwood at Science Based Medicine in his description of EBM's inability to evaluate implausible health claims in the arena of complementary and alternative medicine (CAM).

Here's the listing of Cochrane CAM reviews. Few if any of the treatments covered in these reviews have enough scientific plausibility or prior evidence to warrant a share of our research funds, let alone a systematic review. But the Cochrane reviewers refuse to call a nutty idea a nutty idea. Submit the wackiest fairy dust treatment you can think of to the folks at Cochrane for review. The worst they're liable to say about it is that there's not enough evidence out there to know whether or not it's a valid treatment, so research (or more research) is warranted.

So here are a few examples.

Homeopathic treatment for influenza. Promising, say the reviewers, although more research needed.

Homeopathic treatment for asthma. RCTs needed, of course.

Acupuncture to enhance in vitro fertilization. It works, declare the Cochrane reviewers. But, they say, more research is needed.

Acupuncture for acute stroke (and the mechanism doesn't seem to matter to the reviewers, evidenced by the fact that they lumped hemorrhagic and ischemic strokes, conditions with entirely different mechanisms, together). The reviewers are not sure whether it works and recommend larger and more “methodologically sound” studies.

Acupuncture for vascular dementia. RCTs are urgently needed, say the reviewers!

EBM has become an enabler of pseudoscience, and that's its grand failure. Atwood explains why in the above link and multiple other posts linked there, and I summarize the reasons briefly here. That's the fatal flaw Richard Smith and DB didn't mention in their otherwise informative and accurate criticisms of EBM.

Rabies Milwaukee protocol

The protocol and related resources can be accessed here. The protocol is complicated and counterintuitive. It relies on principles which differ from those of general critical illness. If faced with a potential case, familiarity with the protocol will help in deciding whether you have the resources to implement it in your own hospital or whether the patient should be shipped.

And here's a recent report in MMWR of a case reported in Virginia, following international travel.

Ictal asystole---when the brain stops the heart

It's an experiment of nature in the field of brain-heart medicine. It can result from a seizure in a patient with normal heart function. Via PACE.

Wednesday, December 15, 2010

Work hour limits shouldn't be just for residents

Old farts need rest too, says Bob Wachter in a guest post at Kevin MD, citing a recent study of radiologists (both residents and attendings) showing deteriorating accuracy with increased fatigue.

Resident fatigue was the initial focus of the patient safety movement and even predated the IOM report by over a decade, as a result of the Libby Zion case. Have initiatives to address physician fatigue improved patient outcomes? Although we can't answer that question with data I'm sure the anecdotal evidence would be impressive, as any of us who trained in the pre-Zion era could attest.

I trained in the days of every third and every other night call when night float was not part of the lexicon. On your call day you went to work early and worked 36 hours straight. Maybe longer. You got home late on your post call day so your spouse could watch you sleep. You occasionally fell asleep while interviewing your patient---typically your sixth admission of the night, at 3 AM. The next day a malignant attending might chide you on the sloppy appearance of your hand written H and P. You didn't dare point out that you were just too tired. Too tired? they might ask incredulously. These anecdotes are comical. But what about mistakes with consequences for patients? How many of us really want to know?

Well, for residents, at least, things have changed. We've gone to the opposite extreme, perhaps even too far. The latest proposal is to require naps. It seemed silly to me in the first grade. I would welcome it now.

If efforts to reduce resident fatigue haven't meaningfully impacted patient safety maybe it's because of the unintended consequences or because no attempt has been made to address the equally important issue of fatigue on the part of practicing physicians. But as Bob points out this would be next to impossible to implement for those who practice in the eat-what-you-kill world of fee for service reimbursement.

So (and my conflict of interest here should be obvious) maybe hospitalists should be the next group to target. After all, many of them are treated as advanced residents anyway, and most are not paid by RVUs. For that to gain traction organized hospital medicine would need to set a standard. This might be a worthwhile project for the Society of Hospital Medicine.

Whither evidence based medicine?

It's been 18 years now since evidence based medicine (EBM) became a catch phrase. Richard Smith, a former editor for BMJ, reflects on the distortions and unintended consequences of EBM as a “movement” in this month's The Scientist.

David Sackett and others once tried, in their classic BMJ editorial, to explain what EBM is and what it isn't. But, 18 years later, among academics, policy makers and practicing physicians alike it seems anything but clear what it is and what it isn't. There is a pure notion of EBM out there somewhere, perhaps best stated in the opening sentence of the Sackett editorial---“It's about integrating individual clinical expertise and the best external evidence.” But in real world application it's all over the place as Smith explains.

That's why it's folly to suppose that some central authority can dictate “what works and what doesn't work” for every patient and somehow homogenize the Dartmouth Atlas maps by bringing doctors into “compliance with EBM.”

More from DB's Med Rants.

Hand washing compliance: real quality or just another performance metric?

First a disclaimer. I am a strong proponent of hand hygiene. I strive to be 100% compliant where I work. But the skeptical side of me sees flaws in the hand hygiene dogma. Tracking hand cleaning rates as a performance metric (not a JC or CMS measure yet but being done at many institutions) may have unintended consequences. Moreover, the claim that hand washing keeps hospitalized patients safe from infections is simply not true. Allow me to explain. But first watch this video on hand hygiene at Vanderbilt, a leader in quality and safety.

It's clear the folks at Vanderbilt are passionate about hand hygiene but how effective is the measure, really? Watch the video closely and keep in mind evidence cited by the CDC in its guidelines that a 15 second hand wash is needed, as well as evidence that pathogens can be transmitted via the stethoscope. The stethoscope, perhaps more than the hands, is a major point of physical contact with the patient.

None of the people seen on this video cleaned their stethoscopes. And while it was difficult to tell how long they took to wash their hands it appeared to be well under 15 seconds in most instances.

Making the rate of hand washing the sole metric (91% at Vandy as reported in this video) may distract attention from equally or more important measures such as the length of the hand washing activity and the rate of stethoscope cleaning.