Sunday, June 30, 2013

Is performance a good surrogate for quality?

No, but a casual glance at this paper might give that impression. It deserves a more careful read. From the article:
Importance Federal efforts about public reporting and quality improvement programs for hospitals have focused primarily on a small number of medical conditions. Whether performance on these conditions accurately predicts the quality of broader hospital care is unknown...
Methods Using national Medicare data, we compared hospital performance at 2322 US acute care hospitals on 30-day risk-adjusted mortality, aggregated across the 3 publicly reported conditions (acute myocardial infarction, congestive heart failure, and pneumonia), with performance on a composite risk-adjusted mortality rate across 9 other common medical conditions, a composite mortality rate across 10 surgical conditions, and both composites combined...
Results..Hospitals in the top quartile of performance on publicly reported conditions had a 3.6% lower absolute risk-adjusted mortality rate on the combined medical-surgical composite than those in the bottom quartile (9.4% vs 13.0%; P less than .001). These top performers on publicly reported conditions had 5 times greater odds of being in the top quartile on the overall combined composite risk-adjusted mortality rate (odds ratio [OR], 5.3; 95% CI, 4.3-6.5)...
Conclusions and Relevance Hospital performance on publicly reported conditions can potentially be used as a signal of overall hospital mortality rates.

You can read more here from Medpage Today.

What is important to understand is that the mortality composites were for conditions OTHER than the three publicly reported ones. The connection between performance measures, real quality and outcomes is intangible and there is no warrant from the findings of this study for a direct cause and effect relationship.

What might the intangible factors be? For one thing, hospitals that score high on performance measures may be more resource rich and better staffed. After all it takes considerable time and effort to dot the i's and cross the t's in the medical record in order to score well. Performance is in large part a labor intensive medical record game. Other factors might be that a high scoring hospital has more hospitalists, more robust systems in place or a stronger culture of excellence. We really don't know, do we?

So, my final take...

What we knew before this study:

Performance does not equal quality.

Although performance measures were implemented to incent adherence to evidence based treatments they led to unintended consequences and have, for the most part, not been proven to benefit patients.

What we know in light of this study:

While there is no direct cause and effect relationship between performance and quality, both may be influenced in like manner by poorly understood external factors.

Saturday, June 29, 2013

Aspirin and plavix together: dual antiplatelet therapy for early treatment of minor stroke or high risk TIA

This recently came out in NEJM and everyone's talking about it:

In a randomized, double-blind, placebo-controlled trial conducted at 114 centers in China, we randomly assigned 5170 patients within 24 hours after the onset of minor ischemic stroke or high-risk TIA to combination therapy with clopidogrel and aspirin (clopidogrel at an initial dose of 300 mg, followed by 75 mg per day for 90 days, plus aspirin at a dose of 75 mg per day for the first 21 days) or to placebo plus aspirin (75 mg per day for 90 days). All participants received open-label aspirin at a clinician-determined dose of 75 to 300 mg on day 1...

Stroke occurred in 8.2% of patients in the clopidogrel–aspirin group, as compared with 11.7% of those in the aspirin group (hazard ratio, 0.68; 95% confidence interval, 0.57 to 0.81; P less than 0.001). Moderate or severe hemorrhage occurred in seven patients (0.3%) in the clopidogrel–aspirin group and in eight (0.3%) in the aspirin group (P=0.73); the rate of hemorrhagic stroke was 0.3% in each group.

Among patients with TIA or minor stroke who can be treated within 24 hours after the onset of symptoms, the combination of clopidogrel and aspirin is superior to aspirin alone for reducing the risk of stroke in the first 90 days and does not increase the risk of hemorrhage.

Here's more from Medpage Today.

Some key points:

Prior studies failed to show benefits of dual antiplatelet therapy but they looked at different patient populations (denser and more catastrophic strokes) and had different designs (later treatment).

This study (CHANCE) looked at a more narrowly defined population. It also treated patients earlier (within 24 hours with nearly half within 12 hours). That is important because the event rate was higher early on, particularly in the first two days.

Patient selection was based on the NIH stroke scale for stroke patients (had to be 3 or less) and the ABCD score for TIA patients (had to be 4 or above). Other important exclusions are described in the body of the paper which is available as free full text at the link above.

It is uncertain how well the conclusions of this study, based on a Chinese population (with different risk factors and possibly different genetically programmed responses to Plavix), apply to patients in the US, where a similar trial is getting underway.

Thursday, June 27, 2013

Metformin versus sulfonylureas and macrovascular disease

From a recently published cohort study:

Patients: Veterans who initiated metformin or sulfonylurea therapy for diabetes. Patients with chronic kidney disease or serious medical illness were excluded.
Measurements: Composite outcome of hospitalization for acute myocardial infarction or stroke, or death...
...crude rates of the composite outcome were 18.2 per 1000 person-years in sulfonylurea users and 10.4 per 1000 person-years in metformin users (adjusted incidence rate difference, 2.2 [95% CI, 1.4 to 3.0] more CVD events with sulfonylureas per 1000 person-years; adjusted hazard ratio [aHR], 1.21 [CI, 1.13 to 1.30]). Results were consistent for both glyburide (aHR, 1.26 [CI, 1.16 to 1.37]) and glipizide (aHR, 1.15 [CI, 1.06 to 1.26]) in subgroups by CVD history, age, body mass index, and albuminuria; in a propensity score–matched cohort analysis; and in sensitivity analyses.

According to the authors and the editors it cannot be said whether metformin confers macrovascular benefit, sulfonylureas cause macrovascular harm or both. In the 1970s the University Group Diabetes Program (UGDP) study showed macrovascular harm attributable to sulfonylurea therapy and the class of agents has since carried a black box warning. The UKPDS also suggested superior outcomes with metformin including specifically the macrovascular outcome of stroke.

The authors of the cohort study cite other prior research suggesting superiority of metformin, particularly for macrovascular outcomes, despite similarity in glycemic control compared to other agents.

The emerging picture is that for microvascular disease the more you lower blood sugar the better, no matter how. For macrovascular disease this does not hold true in simple fashion. If lowering blood sugar does matter, it matters how and with what.

Wednesday, June 26, 2013

Cardiac risk after pneumonia

There is more and more coming out about this. Recently the green journal published Pneumonia: An Arrhythmogenic Disease?

Recent studies suggest that there is an increase in cardiovascular disease after pneumonia; however, there is little information on cardiac arrhythmias after pneumonia. The aims of this study were to assess the incidence of, and examine risk factors for, cardiac arrhythmias after hospitalization for pneumonia.
We conducted a national cohort study using Department of Veterans Affairs administrative data including patients aged greater than or equal to 65 years hospitalized with pneumonia in fiscal years 2002-2007, receiving antibiotics within 48 hours of admission, having no prior diagnosis of a cardiac arrhythmia, and having at least 1 year of Veterans Affairs care...
We identified 32,689 patients who met the inclusion criteria. Of these, 3919 (12%) had a new diagnosis of cardiac arrhythmia within 90 days of admission. Variables significantly associated with increased risk of cardiac arrhythmia included increasing age, history of congestive heart failure, and a need for mechanical ventilation or vasopressors. Beta-blocker use was associated with a decreased incidence of events.
An important number of patients have new cardiac arrhythmia during and after hospitalization for pneumonia. Additional research is needed to determine whether use of cardioprotective medications will improve outcomes for patients hospitalized with pneumonia.  At-risk patients hospitalized with pneumonia should be monitored for cardiac arrhythmias during the hospitalization.

(My italics above).

Additional reading:
Aspirin to decrease adverse cardiac events in pneumonia patients.

Sudden cardiac death in pneumonia and other infections.

Sunday, June 23, 2013

More evidence of harm related to proton pump inhibitors

This study found a 1 year increase in mortality in discharged elders, attributable mainly to high dose PPI use. PPIs are great drugs but they must be used discriminately. Overuse may be driven by the belief that these drugs are benign, consumer demand and poorly conceived care bundles resulting in inappropriate prescribing for “GI prophylaxis.” The mechanisms of harm are gradually being discovered. See here.  

Saturday, June 22, 2013

Locum hospitalists: pros and cons

Though hospitalist programs have long relied heavily on locums and continue to do so, increasing controversy surrounds the practice according to this article in Today's Hospitalist. Staffing company and hospitalist program leaders interviewed for the article say the difficult issues are the high cost of hiring locum doctors, instability of group culture and perceived quality problems. Quality issues are not necessarily inherent in the locum doctor, as many excellent hospitalists see locum work as an opportunity for flexibility and better compensation. But even great locum docs may look inferior on various metrics (artificial quality surrogates) because they lack a sense of group ownership and are not under institutional performance incentives. Part of the appeal of locum tenens work may be a sense of freedom from performance pressure.

If hospitalist groups intend to rely less on staffing companies as some leaders claim the trade off is to look to other sources of temporary help such as within-the-group moonlighting. No matter how you slice it staffing shortages remain critical and there's no sign of relief, making me wonder if last year's downtick in hospitalist compensation was an anomaly.

Friday, June 21, 2013

Cuba's health care system

Good on prevention, not so good if you get sick. The authors of this NEJM piece seem to like it but don't romanticize it, they say. OK. From the article:

The system is not designed for consumer choice or individual initiatives. There is no alternative, private-payer health system. Physicians get government benefits such as housing and food subsidies, but they are paid only about $20 per month. Their education is free, and they are respected, but they are unlikely to attain personal wealth. Cuba is a country where 80% of the citizens work for the government, and the government manages the budgets...
A nephrologist in Cienfuegos, 160 miles south of Havana, lists 77 patients on dialysis in the province, which on a population basis is about 40% of the current U.S. rate — similar to what the U.S. rate was in 1985. A neurologist reports that his hospital got a CT scanner only 12 years ago. U.S. students who are enrolled in a Cuban medical school say that operating rooms run quickly and efficiently but with very little technology. Access to information through the Internet is minimal. One medical student reports being limited to 30 minutes per week of dial-up access.

The authors suggest that's the result of the U.S. Economic embargo.

Tuesday, June 18, 2013

Bicuspid aortic valve

----think beyond the valve. It's really a disease of the entire aorta. From the review:

A more recent study has looked at the incidence of aortic complications in 416 BAV patients (mean and median age 35 years, range less than 1–89) [32]. Incidence of aortic dissection was found to be 1.5% in all patients regardless of the progression of BAV; however this increased markedly in patients aged 50 or older at baseline to 17.4% and even more in those found to have aneurysm formation at baseline to (44.9%). 25-year rate for aortic surgery was 25% and there was a significant burden of progression of disease to cause aortic dissection with 49 of the 384 patients without baseline aneurysms developing them during followup, giving an age-adjusted relative risk of 86.2 and an incidence of 84.9 cases per 10000 patient-years.

Monday, June 17, 2013

From hospitalist as clinician to hospitalist as off-loader: how we've changed!

I should know because I was involved in the movement nearly from the beginning. Back then the distinctive was that hospitalists were doctors who, having ascended a steep learning curve from practicing exclusively inpatient medicine, had a special degree of expertise and skill to meet the challenges of caring for the incredibly ill and complex patients who populate the hospital. In short, we were (once) valued for our clinical skills.

Right in line with this was the understanding that effective use of such skills would require enough time to spend with each patient. After all being a good clinician takes time. Practicing evidence based medicine takes time. This led early programs to define who and what they were clinically and to carefully structure coverage arrangements with appropriate restrictions. No one in their wildest imagination had any notion of taking over the hospital or admitting for everyone. At least not in the beginning. But hospitalist leaders started planting the seeds of that idea a few years ago.

I saw trouble brewing then and have blogged my concern many times since. I once made fun of the trend I observed as the hospitalist role was devalued and transitioned from expert clinician to utility player:

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)

So I recently found another published example of the trend in a recent issue of Today's Hospitalist: Universal admitters: hospitalists' new identity? The article cites the growing trend of hospitalists admitting nearly all comers regardless of the clinical need. This according to the article off-loads the specialists, expands the hospitalist's niche and serves the business needs of the institution. Several program leaders were interviewed. To their credit they paid lip service to the concerns about hospitalists becoming overburdened or taking on duties outside the scope of their training. Ultimately, though, to my disappointment, they seemed to one degree or another complicit with the new agenda.

For example:

"The greatest difficulty for hospitalist groups is not recognizing the value in being on the front lines of caring for the majority of patients," Dr. Bossard says. "Going into meetings with the idea that we're going to try to protect our turf and reduce the number of patients we carry by establishing rules to protect us won't work."

I have a little different view of protecting turf. If only the specialty services would protect their own turf! But they seem eager to give it up while hospitalists at the organizational level seem all to willing to capitulate and take it on.

Clinical skill has all but vanished as the raison d'etre for hospitalists. And in the article it's missing from the discussion of all the things the hospitalist “brings to the table,” for example computer skills:

Another factor driving hospitalists toward more admissions: Subspecialists are frustrated with electronic medical records. It makes sense to hand off the computer work associated with being the admitting physician to hospitalists, who log in much more screen time.

Though mentioned as an aside that paragraph says a lot about the changing agenda: the move toward hospitalist as secretary, H&P provider and off-loader.

Sunday, June 16, 2013

Guillain-Barré review

Once an open access journal, American Family Physician for the past few years has been behind heavy access controls. Every once in a while a really good review shows up there. Recently they published this article on Guillain-Barré syndrome. It is worth reading in the full text if you can get it.

Some points of interest:

Though the varied spectrum of GBS has enough important features in common to discuss for practical purposes as a single entity it is probably best thought of as a group of diseases reflecting the different immune responses, different targets of immune injury (myelin vs axon), differences in epidemiology and the clinical variants.

Pain and autonomic instability are very common features, I think more than generally appreciated. So maybe you've got a patient presenting with funny unexplained pains and vital signs all over the map. Maybe with multiple preceding ER visits. Think of it!

Symptoms are bilateral and tend to be symmetric but may not be perfectly so at all points in the clinical course.

Reflexes have to be decreased but not necessarily absent at all points in the clinical course.

The epidemiology is complex. While often discussed in the context of Campylobacter jejuni, that infection is associated with a variant of GBS (axonal) not typical in the U.S. A variety of illnesses may precede GBS.

There are two arms of treatment: supportive and disease specific.

Supportive care addresses DVT prophylaxis, pain control, skin care, bowel and bladder issues, physical therapy, cardiac and vital sign monitoring and respiratory care. Multiple respiratory parameters can be monitored. The article lists, among other signs, forced vital capacity less than 20 mL/kg, maximum inspiratory pressure less than 30 cmH2O, and maximum expiratory pressure less than 40 cmH2O as risk factors for respiratory failure. The presence of any of the three signs, according to Up to Date, is an indication for urgent intubation and mechanical ventilation.

Disease specific modalities are plasma exchange and IVIG. One or the other is used, not both. Is one preferred over the other? The review article says plasma exchange is “first line.” The Up to Date section on treatment of GBS says the treatments are equivalent, and that IVIG tends to be favored due to availability and ease of use.

Saturday, June 15, 2013

Is the state medical board watching your on line behavior?

A recent paper reports a survey of state medical boards as to how they would respond to hypothetical scenarios:

 High consensus was defined as more than 75% of respondents indicating that investigation was “likely” or “very likely,” moderate consensus as 50% to 75% indicating this, and low consensus as fewer than 50% indicating this.
Four online vignettes demonstrated high consensus: Citing misleading information about clinical outcomes (81%; 39/48), using patient images without consent (79%; 38/48), misrepresenting credentials (77%; 37/48), and inappropriately contacting patients (77%; 37/48). Three demonstrated moderate consensus for investigation: depicting alcohol intoxication (73%; 35/48), violating patient confidentiality (65%; 31/48), and using discriminatory speech (60%; 29/48). Three demonstrated low consensus: using derogatory speech toward patients (46%; 22/48), showing alcohol use without intoxication (40%; 19/48), and providing clinical narratives without violation of confidentiality (16%; 7/48).

Friday, June 14, 2013

Electrocardiographic abnormalities in pneumonia

In this study transient ECG abnormalities were seen around the time of the episode of pneumonia. Some resembled those seen in PE.

Thursday, June 13, 2013

PFO closure after cryptogenic stroke: evidence still lacking

---after two more trials published in NEJM here and here.

There were problems with the data, leaving open the possibility of benefit. But it remains unproven. Perspective here via Medpage Today.

Tuesday, June 11, 2013

Heart failure with preserved ejection fraction (HFPEF)

---was formerly known as diastolic heart failure. A recent paper was published in the American Journal of Medicine: Contemporary Prevalence and Correlates of Incident Heart Failure with Preserved Ejection Fraction:

We assessed the prevalence of preserved left ventricular ejection fraction in patients with incident heart failure and differences in the demographic and clinical characteristics that may differentiate patients presenting with heart failure with preserved versus reduced left ventricular ejection fraction.
We identified all patients with newly diagnosed heart failure between 2005 and 2008 from 4 sites in the Cardiovascular Research Network on the basis of hospital discharge and ambulatory visit diagnoses, and assigned a category of preserved, borderline, or reduced left ventricular ejection fraction using data from electronic databases and chart review.
We identified 11,994 patients with incident heart failure; of these, 6210 (51.8%) had preserved left ventricular ejection fraction, 1870 (15.6%) had borderline systolic dysfunction, and 3914 (32.6%) had reduced left ventricular ejection fraction.

The authors list these key points:

•Heart failure with preserved left ventricular ejection fraction is the most common form of the heart failure syndrome among patients newly presenting with this condition.
•Women and older adults are especially affected.
•Patients with ejection fractions less than 40% comprise less than one third of those with incident heart failure.
•Evidence-based treatment strategies apply to only a minority of patients with newly diagnosed heart failure.

What does that last statement mean? Treatments have been validated by high level evidence for long term clinical outcomes (for the most part treatments with life prolonging neurohumoral antagonists) only for patients with clearly reduced ejection fractions. But those patients made up less than one third of the incident heart failure cases in this database. There's a commonly held maxim that “there is no evidence-based treatment available for HFPEF.” That's not exactly true. While there's no generalized treatment for the disease per se there are treatments for its manifestations (diuretics for pulmonary edema, rate controlling medications for atrial fibrillation) as well as its associated risk factors and comorbidities. These are, indeed, evidence-based and they treat the patient more than the disease.

Monday, June 10, 2013

Point of care echo in the assessment and management of the critically ill

According to this review it is emerging as a core skill for critical care docs. Will it be such for hospitalists??

From the review:

In recent years, several focused echocardiography protocols have been introduced [12]. These studies can usually be carried out by novice operators after a modest amount of training.

There's hope!

Sunday, June 09, 2013

Cutaneous signs of systemic disease: necrolytic migratory erythema

A case report and brief review of glucagonoma is presented here along with a discussion of a few other entities that can produce the same dermatosis. Free full text.

Friday, June 07, 2013

Lyme disease misconceptions

Misconceptions about Lyme disease abound and have fueled politicized debates and a good deal of quackery. A recent piece in the green journal (free full text) provides some clarity and includes a list of bullet points about the misconceptions. Here are a few:

“'Blood tests are unreliable with many negatives in patients who really have Lyme disease'”
“Just as with all antibody-based testing, these are often negative very early before the antibody response develops (less than 4-6 weeks). They are rarely if ever negative in later disease.”

“'Antibiotics make blood tests negative during treatment'”
“There is no evidence that this happens and no biologic reason it would.”

“'Lyme disease is a clinical diagnosis that should be made based on a list of symptoms'”
“No clinical features, except erythema migrans or possibly bilateral facial nerve palsy—in the appropriate context—provide sufficient specificity or positive predictive value. Laboratory confirmation is essential except with erythema migrans.”

“'B. burgdorferi infection is potentially lethal'”
“Although Lyme disease can cause heart or brain abnormalities, there have been remarkably few—if any—deaths attributable to this infection”

“'If, following treatment, symptoms persist, or serologic testing remains positive, additional treatment is required'”
“Multiple well-performed studies demonstrate that recommended treatment courses cure this infection. Retreatment is necessary occasionally, but not frequently.”

Now as readers know I'm not a fan of the Institute of Medicine (IOM) but they are quoted in the article and I happen to agree with what they said:

...strong emotions, mistrust, and a game of blaming others who are not aligned with one's views” have resulted in a heated and politicized debate. A number of factors have contributed to this “debate”—perhaps not the least of which is a tension between the concept of evidence-based medicine and medicine's historical inductive approach from anecdotal observation. This tension is reflected in 2 frequently repeated, interrelated assertions—that laboratory testing for Lyme disease is unreliable and that the disease should be defined “clinically”—meaning syndromically.

Thursday, June 06, 2013

AVNRT update

Much of this review is applicable to the EP lab although it contains some updated information on the surface ECG and how it can help in differentiating AVNRT from other SVTs.

Wednesday, June 05, 2013

Metformin use and the prognostic value of lactate in sepsis

From an article in the American Journal of Emergency Medicine:

A total of 1947 ED patients were enrolled; 192 (10%) were taking metformin; 305 (16%) died within 28 days. Metformin users had higher median lactate levels than nonusers (2.2 mmol/L [interquartile range, 1.6-3.2] vs 1.9 mmol/L [interquartile range, 1.3-2.8]) and a higher, although nonsignificant, prevalence of hyperlactatemia (lactate greater than or equal to 4.0 mmol/L) (17% vs 13%) (P = .17). In multivariate analysis (reference group nonmetformin users, lactate less than 2.0 mmol/L), hyperlactatemia was associated with an increased adjusted 28-day mortality risk among nonmetformin users (odds ratio [OR], 3.18; P less than .01) but not among metformin users (OR, 0.54; P = .33). In addition, nonmetformin users had a higher adjusted mortality risk than metformin users (OR, 2.49; P less than .01). These differences remained significant when only diabetic patients were analyzed.
In this study of adult ED patients with suspected sepsis, metformin users had slightly higher median lactate levels and prevalence of hyperlactatemia. However, hyperlactatemia did not predict an increased mortality risk in patients taking metformin.

Tuesday, June 04, 2013

Kayexalate and GI injury

According to a recent systematic review:

Thirty reports describing 58 cases (41 preparations containing sorbitol and 17 preparations without sorbitol) of adverse events were identified. The colon was the most common site of injury (n=44; 76%), and transmural necrosis (n=36; 62%) was the most common histopathologic lesion reported. Mortality was reported in 33% of these cases due to gastrointestinal injury.
Sodium polystyrene sulfonate use, both with and without sorbitol, may be associated with fatal gastrointestinal injury. Physicians must be cognizant of the risk of these adverse events when prescribing this therapy for the management of hyperkalemia.

Monday, June 03, 2013

Routine use of PEEP in mechanical ventilation

The only proven indication for PEEP is hypoxemia that is refractory to reasonable fractions of inspired oxygen. Yet, “five of PEEP” has become a default setting in mechanical ventilation no matter what the indication due to physiologic rationale and tradition. If cardiac decompensation (cardiogenic pulmonary edema) is in the loop of respiratory failure the physiologic rationale becomes even stronger because positive pressure applied anytime during the breathing cycle decreases afterload and has other hemodynamic benefits. That was the subject of this recent paper. The evidence is scant but PEEP appears to be safe and may offer clinical benefit when cardiac decompensation causes or complicates respiratory failure.

Saturday, June 01, 2013

Long QT: what hospitalists need to know

Here is an update in the American Journal of Emergency Medicine. Both the congenital and acquired forms are discussed. For congenital LQTS beta blockers are mentioned as first line therapy. The general principle guiding this recommendation is based on two different types of triggers for Torsades in patients with long QT: sympathetic surge (generally seen in the congenital form) and increased cardiac cycle length in the form of pause or bradycardia (the usual trigger in acquired cases). Overlap exists, however. Some acquired cases represent a mutation with weak expression, the phenotype manifesting for the first time in the face of an external factor such as an electrolyte disturbance or drug effect. And while the majority of genotypes respond to beta blockers some may not. The usual role of the hospitalist is to recognize abnormal repolarization in its various forms and address external triggers such as drugs and electrolyte disturbances, and know when to call the electrophysiologist.