Thursday, December 31, 2009
Safety means minimizing harm that comes to patients in the course of their care. Although the patient safety movement got started as a result of events following the Libby Zion case in the 1980s the release of the Institute of Medicine Report on patient safety a decade ago was a defining moment.
Hospitalists are positioned to be heavily involved in both quality and safety. It seems appropriate at the end of the decade to ask: “How are we doing?”
Although the news about quality is not all bad the quality movement overall has largely failed due to confusion between quality and performance. A focus on performance is more about institutional narcissism than about the well being of the patient. That said, let's look at some individual quality measures.
Acute coronary syndrome:
ASA: Evidence based and robust, but already widely done. Effectiveness of making it a performance measure doubtful.
Beta blockers, ACEI/ARBs: Evidence based, but the effect of the performance measure is probably weak. Hospital providers can play for the report card by starting a small dose at discharge, which may never get titrated to goal in the clinic.
Smoking cessation education: Weak at best. Although hospital providers could seize a unique opportunity to change the patient's lifestyle that would require spending considerable time with the patient. Instead, in order to play for the report card, busy hospital personnel are more likely to merely print off a few pages of information and hand it to the patient as s/he's on the way out the door. This and other performance measures may distract from non-performance but equally important measures such as statin therapy.
Time to reperfusion: Hospitals are scrambling to get their door to balloon times down. This is a strong measure that may be the one success story among ACS performance measures.
Discharge instructions: Same as for smoking cessation above. In the studies which showed discharge instructions to improve outcomes a specialty nurse sat down and spent an hour or two with the patient and family. Hospitals, concerned about bed control and early discharge, don't feel they have time to do this and don't have to to play for the report card.
Evaluation for left ventricular systolic dysfunction: This is an important assessment which helps evaluate patients' candidacy for multiple therapies. Since most of these take place after the patient is discharged it makes little difference whether the test is done that admission. Moreover, it does no good at all if the results are not followed up appropriately in clinic. So, making it a hospital performance measure probably has little effect.
ACEI/ARB therapy: Just as for MI, these agents are likely to be prescribed in a small dose to satisfy the measure, then never get titrated to goal in clinic. Providers who are uncomfortable prescribing an ACEI or ARB can also satisfy the measure by finding and documenting any of several soft “contraindications” such as low blood pressure or elevated creatinine.
The heart failure performance measures may distract from more important therapies which are not performance measures such as beta blockers and devices.
It's not hard to imagine then why the OPTIMIZE database found no evidence of effectiveness of heart failure performance measures.
Pneumococcal vaccination: The pneumococcal vaccine approved for adults is weak and largely ineffective.
Blood cultures in the emergency room: This is an important measure for many patients with pneumonia. The evidence suggests it is not helpful in some others. Supplanting clinical judgment with a performance measure for this test is inappropriate.
Smoking cessation: A weak measure. See comments above.
Time to antibiotic: This measure has been found repeatedly to result in non-evidence based administration of antibiotics without clinical benefit and possible harm. It has been removed from the guidelines. Nevertheless it persists as a performance measure.
Appropriateness of antibiotics: Efforts to encourage guideline adherence in this area is laudable. However, performance measures may lag behind the latest evidence. What's more it may be all too easy to adjust the documentation to fit the antibiotic you gave rather than the other way around.
Influenza vaccination: This is evidence based and a much stronger measure than pneumococcal vaccination.
Evaluation of our progress in advancing the cause of patient safety has been controversial. Bob Wachter gave the movement a B-. I was not so optimistic. A leading consumer group characterized it as a public policy failure. I tried to resolve the controversy here and elsewhere.
According to Bob Wachter one of the problems with patient safety is that we've taken the no blame idea too far. I countered that we actually have, as an unintended result of the IOM report, a culture of blame which has damaged the patient safety movement and cited evidence that this culture of blame has damaged the cause of transparency. This year I also expressed the opinion that the idea of “accountability versus no blame” was a false dichotomy.
Top 10 issues in hospital medicine for 2009, issue 2: hospitalist job descriptions and career satisfaction
These hospitalists were overworked and, being at the beck and call of administration, constantly interrupted. Although the administrator was a physician he didn't seem very sympathetic to their professional needs.
Building an excellent hospitalist program requires hiring and retaining excellent (career) hospitalists. To attract and retain excellent hospitalists a program must offer professional satisfaction. Professional satisfaction depends on the job description. Therein lies the problem. The increasingly amorphous description of the hospitalist job is badly in need of definition as hospitalists are increasingly viewed as utility players and business solutions responsible for an increasing array of non clinical tasks.
The Society of Hospital Medicine urgently needs to address this problem. So far, unfortunately, they have taken little initiative in doing so, at least from what I've been able to observe.
To supporters and haters alike, agonizing and anxiously awaiting the results of the battery of tests being given to Limbaugh by his Hawaii doctors, worry not! Bill Shireman at the Huffington Post has nailed the diagnosis. Is it angina? GERD? Musculoskeletal pain? NO!
You see, Shireman explains, Limbaugh has two addictions: not only pain medication but also the adoration of his fans. He can only service this latter addiction while he's in studio doing his daily radio show. Ergo:
That he experienced chest pains while on "vacation" just illustrates the point. I can't imagine any experience more disconcerting to a socially fearful celebrity than an unstructured Hawaiian vacation. People who feel they are fundamentally inauthentic or undeserving are bound to unhinge a bit when left alone with their self-awareness.
Brilliant! I hope someone lets his doctors in Hawaii know before those greedy cardiologists do all their unnecessary procedures.
A paper in the Archives of Internal Medicine this year purported that hospitalists improve quality, but suffered from confusion on quality and performance. Dr. Robert Centor gave us a dose of healthy skepticism.
This comes up every year and 2009 was no exception. Despite new developments and continued attacks on the Surviving Sepsis Campaign the guidelines, I believe, are still reliable. Here's what caught my attention this year:
Early goal directed therapy
A meta-analysis confirmed that early, but not late, goal directed therapy works.
Despite public controversy about EGDT evidence continues to mount in favor of the protocol and the pathophysiologic rationale is compelling. EGDT is a package of interventions, and one of the controversies is whether the entire package is necessary or whether some component interventions can be eliminated. That question is the subject of a NIH trial just getting underway.
A Bayesian analysis was critical of EGDT but did not take into account the appropriate prior evidence.
Point of care measurement in the ER is useful.
Choice of pressor agent
Although evidence tends to favor norepinephrine the issue remains controversial, and optimal use requires taking into account the individual patient's hemodynamic status. I cited a new paper and a helpful topic review here.
Activated protein C
Although this agent remains controversial I didn't find a lot that was new in 2009. There was one interesting paper about the appropriateness of the product labeling, pointing out that the contraindications do not include all the patient exclusion criteria in PROWESS. If the exclusions in PROWESS (which encompass both the contraindications and the “warnings and precautions” in the labeling) are followed in clinical use the drug seems to perform better from a safety standpoint.
After mounting negative evidence concerning intensive glycemic control in critical illness this year saw publication of NICE-SUGAR which failed to show benefit. What do we know at the end of 2009? We know we shouldn't ignore the blood glucose in septic patients but we don't know the optimal target. An editorial put it this way:
Despite these caveats, we think it is important to emphasize that the findings of NICE-SUGAR do not justify neglecting glycemic control. Instead, we think that, whatever the mechanisms behind the findings of NICE-SUGAR, there is now a new and more moderate standard of care for glycemic management in the ICU: do not treat hyperglycemia unless the glucose level increases higher than 180 mg/dL; when you do treat hyperglycemia, aim for a target blood glucose concentration between 144 and 180 mg/dL. Until a study can provide level I evidence that a better approach exists, this should remain the standard of care. Such a standard of care also implies that, for example, in patients in the ICU, a glucose level of 243 mg/dL is just as undesirable as a glucose level of 80 mg/dL.
Did CORTICUS, published almost two years ago, close the book on this controversy? Far from it. A systematic review published this year may have swung the pendulum back just a bit. What's really going on? Critical illness related corticosteroid insufficiency (CIRCI) is a real entity. What's more, occasional patients with sepsis may have full blown adrenal insufficiency resembling classic Addisonism. When all is said and done the current guideline recommendation that corticosteroids be reserved for patients who are refractory to pressors seems reasonable. Clinical judgment, however, may dictate other approaches in individual patients. I posted on this topic last July.
Initial antibiotic therapy
Growing evidence shows that covering the spectrum adequately, and doing it early, are important.
Surviving Sepsis Campaign
Promotion of the Surviving Sepsis Guidelines increased adherence and saved lives. This is the first study ever done, as far as I know, on the effect of pharmaceutical industry promotion on hard clinical outcomes.
The field of infectious disease is a moving target and every year brings new concerns for hospitalists. Here's what stood out in 2009:
Gram negative resistance
The big concerns this year were extended spectrum betalactamases (ESBLs), Klebsiella pneumoniae carbapenemases (KPCs which, by the way, are found in a variety of gram negative bacteria, not just Klebsiella) and Acinitobacter. With nothing in the pipeline right now clinicians are turning to tygecycline and our old friends the polymyxins. Polymyxins carry toxixity concerns. Tygecycline's high tissue penetration and consequent low serum levels raise concerns about its use in septic patients. This year I posted updates here, here and here.
What continues to amaze me about serious staphylococcal infections is the nuanced approach that is necessary. “Hang vanco and call it a day” just won't cut it. Here's what stood out this year:
Using a dying drug for all it's worth: consensus statement on vancomycin use.
Will vancomycin be enough? Look at the MIC!
What this all means is treating serious staphylococcal infections is tricky business and demands expertise. A growing literature suggests that ID consultation improves outcomes. What do the ID docs add? They know the appropriate target levels for vancomycin. They know when one of the newer antibiotics (daptomycin, linezolid, etc.) should be used instead of vancomycin. They can help to decide when and how to decolonize. They're more likely to use beta lactams over vancomycin for MSSA. They're more likely to do follow up blood cultures and treat long enough. They take a more aggressive approach to source identification and removal.
Image courtesy of the Missouri Historical Archive.
Wednesday, December 30, 2009
For several years I've been writing about aspects of resuscitation and post-resuscitation care starting with a post on Dr. Gordon Ewy's cardiocerebral resuscitation. Therapeutic hypothermia following cardiac arrest received increased attention this year and I posted on the subject here, here, here and here.
Accumulating evidence on other aspects of post resuscitation care, particularly aggressive coronary angiography and PCI strategies, has led to the notion of the post-cardiac arrest syndrome and post-resuscitation bundle, recently promulgated as an ILCOR consensus statement.
This Medscape Viewpoint piece provides a helpful perspective on the current evidence for the role of emergency PCI in resuscitated patients:
Reynolds' study further supports previously published reports encouraging urgent catheterization for survivors of cardiac arrest regardless of ECG evidence of STEMI. Recent publications also show that therapeutic hypothermia can be used safely in these patients during and after PCI without producing delays in time to balloon inflation.
The significance of this new literature cannot be overstated. If further studies confirm these findings, it would strongly argue for enormous changes in prehospital systems of care. All survivors of primary cardiac arrest would be recommended for immediate transport to hospitals that have the capability of performing urgent PCI in conjunction with therapeutic hypothermia. Based on the current literature, it certainly seems advisable that emergency healthcare practitioners who care for resuscitated victims of primary cardiac arrest should engage in conversations with cardiology consultants and urge them to take an aggressive approach to PCI in these patients.
This new integrated approach to cardiac arrest care will require changes in the way we assess neurologic prognosis after cardiac arrest. We won't be giving up so soon, because assessment of neurologic prognosis will have to be delayed in patients who receive therapeutic hypothermia. Again, from the Medscape article (my emphasis):
The authors determined that improved survival and good outcome were associated with coronary angiography regardless of the presence of new left bundle branch block or STEMI, and also regardless of presenting rhythm or neurologic status immediately after resuscitation.
A nice overview of resuscitation and post-resuscitation care is contained in this review from Current Opinion in Cardiology. According to that review post-resuscitation care should include:
...mild hypothermia (32–348C) for patients in coma after arrest. Urgent cardiac catheterizationand PCI unless contraindicated .
Top 10 issues in hospital medicine for 2009, issue 7: rapid response teams, telemetry and early warning scores
The premise behind rapid response teams (RRTs), that intervention in deteriorating patients should avert cardiac arrests and reduce hospital mortality, is appealing. Yet, despite widespread claims that such teams save lives, extensive research has failed to show that they do.
Why wouldn't it be effective? Maybe because customary use of RRTs is to intervene at the last minute when, in most cases, physiologic deterioration begins hours before the crash. That physiologic deterioration may, in cases such as sepsis, be irreversible in the final moments leading to cardiopulmonary arrest. Having a RRT is not a substitute for clinical vigilance.
The consensus of a panel of speakers at SHM 2009 was that while research evidence doesn't show an impact on meaningful clinical outcomes individual institutions may realize intangible benefits. Nurses love the concept. Rapid response calls may lead to advance health care planning discussions. Perhaps most importantly, rapid response systems can be educational tools. When reviewed systematically they may uncover safety issues and drive process improvement.
The panelists discussed tips for more effective use of RRTs, learned from their own institutions. The process should be structured. Define the patient changes that should trigger a call. Formally review all RRT call records. Define post-RRT care and hand-off procedures. The RRT concept can be extended in novel ways such as proactive rounds on high acuity patients and extending coverage beyond the wards to the ED and other areas of the facility.
Whatever the benefits of the RRT for the individual institution, based on the disappointing literature on RRT outcomes no one can legitimately claim that their institution is saving lives just because they have a RRT. Moreover, there is no research basis for making the RRT a reportable quality measure.
Before leaving the topic of RRTs, a word about Joint Commission. Contrary to popular belief Joint Commission does not require hospitals to have RRTs. Instead they require hospitals to have a structured mechanism in place, specific to the needs of the individual facility, whereby help is summoned for deteriorating patients. In addition hospitals are to monitor outcomes before and after implementation.
Have other means of averting cardiac arrest proven effective? Earlier this year I wrote a post about telemetry monitoring. Telemetry monitoring alerts nursing staff of cardiac arrest in progress, but only rarely prevents cardiac arrest and is over-utilized.
Because physiologic changes which progress to cardiac arrest often unfold over hours and may be irreversible in the moments preceding cardiopulmonary arrest, early warning assessment methods have been used. Perhaps the most popular is the modified early warning score (MEWS) a physiologic scoring tool which, in this study, was highly predictive of the risk of in hospital deterioration. This tool can be used at the time of hospital admission and periodically thereafter for early identification of patients at risk so that intervention can take place before the patient is in crisis.
Finally, of related interest this year was this note about Vanderbilt's in patient sepsis early warning and management system.
Image courtesy of the Missouri Historical archives.
An important NEJM paper published this year reported these findings on discharged Medicare patients:
Almost one fifth (19.6%) of the 11,855,702 Medicare beneficiaries who had been discharged from a hospital were rehospitalized within 30 days, and 34.0% were rehospitalized within 90 days; 67.1% of patients who had been discharged with medical conditions and 51.5% of those who had been discharged after surgical procedures were rehospitalized or died within the first year after discharge.
Although differing methodologies make direct comparison difficult these results appear to be worse than those reported in a similar study during the pre-DRG era 25 years ago. It's reasonable to suspect that the arrival of DRG reimbursement made the problem worse. Suddenly Medicare no longer paid for the hospital care patients got, let alone what they actually needed. Hospitals, in order to survive, quickly responded by discharging patients “quicker and sicker.” According to a JAMA report:
Instability at discharge (important clinical problems usually first occurring prior to discharge) predicted the likelihood of postdischarge deaths. At 90 days postdischarge, 16% of patients discharged unstable were dead vs 10% of patients discharged stable. After the PPS introduction, instability increased primarily among patients discharged home. Prior to the PPS, 10% of patients discharged home were unstable; after the PPS was implemented, 15% were discharged unstable, a 43% relative change.
Although the enactment of DRGs, an ill-conceived and reactionary move against the preceding 20 years of gross Medicare largess was, like the health care reform proposals of today, supposed to make care more efficient, it created a mess. Fortunately, back in the 1980s there were plenty of primary care docs to pick up the pieces.
Now things are different. As Medicare continues its increasingly negative cost incentives the primary care work force has diminished. So we shouldn't be surprised that the recent NEJM paper found this:
In the case of 50.2% of the patients who were rehospitalized within 30 days after a medical discharge to the community, there was no bill for a visit to a physician's office between the time of discharge and rehospitalization.
I remain skeptical about government's ability to improve this situation through external pressure. Meanwhile hospitalists, because their model has aggravated the disconnect between hospital and outpatient care, have an obligation to try and improve discharge transitions. Some programs are trying to lead the way in this effort through initiatives like Project Boost. Unfortunately efforts such as these take resources and may come into conflict with hospitals' business models under the DRG system.
Image courtesy of the Missouri Historical archives.
Top 10 issues in hospital medicine for 2009, issue 10: the rising importance of statin drugs in hospitalized patients
Statins improve the COPD outcomes of mortality, frequency of exacerbations, frequency of mechanical ventilation and frequency of need for emergency care.
Perioperative statin use cuts myocardial ischemic events in half in patients undergoing vascular surgery.
More rapid titration of statin drugs to goal may reduce events.
Acute statin dosing may improve acute coronary syndrome outcomes.
Pleiotropic statin effects improve stroke outcomes.
Statins improve outcomes in sepsis.
Acute statin use in ACS and perioperative statin use represent potential opportunities for intervention. The data for COPD, sepsis and stroke represent comparisons between users and non users. That may be important for hospitalists in deciding which home medications to continue in the hospital when patients are admitted with these conditions.
Don't be surprised in a few years if statin administration, like DVT prophylaxis, shows up on your hospital safety check list.
It's time again to reflect on hospital medicine issues of importance for the past year. In a series of posts over the next couple of days I'll be offering my totally subjective and biased top 10 list, a mixture of clinical and organizational issues. I've been doing this for the last several years. You can view previous entries here.
Monday, December 28, 2009
What does the literature say?
In this retrospective cohort study:
Patients who received IDC had closer blood culture follow-up and better antibiotic selection, and were more likely to have pus or prosthetic material removed. Hospital mortality from SAB was lower in patients who received IDC than in those who did not (13.9% vs. 23.7%; p = 0.05). In multivariate survival analysis, IDC was associated with substantially lower hazard of hospital mortality during SAB (hazard 0.46; p = 0.03).
...In conclusion, IDC is associated with reduced mortality in patients with staphylococcal bacteremia.
In another recent study:
Factors significantly associated with in-hospital mortality in multivariate analysis were ICU admission (OR 5.8, CI 3.5–9.7), MRSA (OR 2.6, CI 1.4–4.9), age greater than or equal to 60 years (OR 2.4, CI 1.4–4.2), a diagnosis of endocarditis (OR 2.8, CI 1.4–5.7), a non-fatal underlying disease/comorbidity according to the McCabe classification (OR 0.2, CI 0.1–0.4), and infectious disease specialist consultation (OR 0.6, CI 0.4–1.0).
These data suggest that outcome of S. aureus bacteremia may be improved by an expert consultation service.
Another study from last year showed a non-statistically significant trend toward decreased treatment failures with ID consultation but an increased rate of detection of complicated infections and some process differences:
Echocardiography (57% vs. 73%; p = .01) and radiographic studies (81% vs. 91%; p = .04) were used more frequently during the period of routine consultation, and infective endocarditis or metastatic infections were diagnosed more frequently (33% vs. 46%; p = .04). All 4 standards of care (removal of intravascular foci of infection, obtaining follow-up blood culture samples, use of parenteral beta-lactam therapy when possible, and administration of greater than or equal to 28 days of therapy for complicated infections) were adhered to more frequently with routine consultation (40% vs. 74%; P less than .001).
So if ID consultation really is producing better outcomes what are the ID docs adding? From the descriptions above, better evaluation and management all the way around, it would appear. Based on the descriptions above, if ID consultation is not readily available at your hospital, be aware of the modalities often overlooked by non-specialists: knowing when vanco is not enough; getting follow up blood cultures; source control; treating long enough; appropriate use of imaging to diagnose endocarditis and other complications; switching to a beta lactam antibiotic if the organism turns out to be MSSA.
At first glance that may seem odd. After all it's often the critics of CAM who want it to be “evidence based”, and although there's one camp of woosters who claim to be evidence based a second camp complains that the methods of EBM are inadequate to evaluate their modalities! I maintain that the second camp happens to be correct, but for the reasons I explained in part 2 and not for the reasons they claim.
Bayesian thinking addresses the problem by evaluating claims of CAM in light of prior science. In a quantitative sense, a scientifically implausible claim like homeopathy would have a prior probability P(A) of virtually zero. Because it is in the numerator of the equation and is infinitesimally small the new (clinical trial) evidence would have to be overwhelming (sufficient to shake the foundations of basic science) in order to support the claim. In common sense terms that means “extraordinary claims require extraordinary proof.” So, for the outlandish claims of woo you can do Bayesian analysis without even bothering with the math. Before I was even aware of Bayesian analysis to evaluate clinical trials I realized this principle intuitively; homeopathy and many of the other forms of woo just made my baloney detector go off.
Bayesian analysis requires appropriate selection of which prior evidence to use. This means including appropriate basic science (which is why systematic reviews and meta-analyses are not enough) and including only that prior information which is applicable. For the extremely implausible claims it's pretty easy. For some questions in mainstream clinical medicine it gets tricky.
An article on sepsis in PloS ONE from May of last year illustrates some of the problems. The authors attempted to apply Bayesian methods to the Surviving Sepsis Campaign bundle and reached this conclusion:
Our results demonstrate that the strength of evidence (statistical and clinical) is weak for all trials, particularly for the Low-Dose Steroid and EGDT trials. It is essential to replicate the results of each of these five clinical trials in confirmatory studies if we want to provide patient care based on scientifically sound evidence.
They further stated that application of only moderately skeptical analysis of prior evidence leads to the conclusion that the evidence is too weak to recommend early goal directed therapy, low dose steroids, intensive insulin therapy, activated protein C or low tidal volume ventilation. At least for EGDT (early goal directed therapy) and low tidal volume ventilation that's a pretty surprising statement and bears examination.
For purposes of this discussion I'll focus on EGDT which, by most, accounts, with the possible exceptions of early antibiotic therapy and source control, is the most robust of the sepsis bundle recommendations. Why did the authors give it such a low rating? I think they made two mistakes in the incorporation of prior evidence. The first is that while they looked at many clinical trials they ignored the more basic evidence that establishes the background and biologic rationale for EGDT, such as the extensive research cited in this paper. The second mistake is that the negative trials on hemodynamic optimization which they cited (see references 40-47) did not examine whether early hemodynamic optimization in the ER is beneficial, which was the premise of EGDT explicitly laid out in Rivers' original paper.
I could go on at length about the use and misuse of EBM, how it fails in the evaluation of CAM modalities and how Bayesian methods can help when properly applied, but enough for now. Perhaps the best way to end is with this perspective from one of Kimball Atwood's posts at Science Based Medicine:
Thus although EBM correctly recognizes that basic science is an insufficient basis for determining the safety and effectiveness of a new medical treatment, it overlooks its necessary place in that exercise.
Tuesday, December 22, 2009
As I mentioned in the previous post a non-quantitative description of Bayesian analysis is that new information must always be interpreted in light of prior information. That's intuitive to most clinicians when they think about clinical trial results. Research findings that disagree with prior research are likely to be viewed with more skepticism. Although Bayesian analysis applies quantitative methods to the incorporation of prior knowledge there are subjective elements. This subjectivity has been the basis for criticism of Bayesian analysis. Such criticism is misinformed because current frequentist methods, with their well known p values and confidence intervals, take no account of prior knowledge and are not as objective as they appear. In fact, assumptions concerning truth in the real clinical world based on a low p value of a single study may be little more than leaps of faith. P values in research reports are analogous to the specificity of laboratory reports in clinical diagnosis as discussed in the previous post. Neither specificity levels nor p values directly address whether a finding is true. In fact both are based on negative assumptions (absence of disease and absence of treatment effect in a clinical trial, respectively). The p value calculation assumes no treatment effect (the null hypothesis). The value itself denotes the probability of obtaining the experimental result if the null hypothesis is true.
To estimate the probability that a research finding reflects generalizable truth one must determine the posterior probability (the probability that the hypothesis is true given the prior knowledge and the new data). This is analogous to the positive predictive value I referenced in part 1 and is ignored by the current frequentist methods of evidence based medicine (EBM). EBM works fairly well most of the time when research results are more or less in line with prior knowledge. But because it ignores prior knowledge, especially basic science, it fails in the evaluation of implausible claims. Consider some of the wooiest of woo, e.g. homeopathy and therapeutic touch. Occasional “research” studies will show positive treatment effects with statistically significant p values. But prior knowledge from basic science, bringing to light the implausibility of such claims, would result in a pre-test probability (prior probability) of virtually zero. Both quantitative Bayesian analysis and plain old common sense tell us that the degree of proof needed to overcome such overwhelming prior odds would have to shake the world of chemistry and physics, requiring a fundamental re-writing of the text books.
Kimball Atwood, in a post from last year in Science Based Medicine, explained EBM's failure to evaluate the claims of complementary and alternative medicine this way:
Evidence-Based Medicine (EBM) is not up to the task of evaluating highly implausible claims. That discussion made the point that EBM favors equivocal clinical trial data over basic science, even if the latter is both firmly established and refutes the clinical claim. It suggested that this failure in calculus is not an indictment of EBM’s originators, but rather was an understandable lapse on their part: it never occurred to them, even as recently as 1990, that EBM would soon be asked to judge contests pitting low powered, bias-prone clinical investigations and reviews against facts of nature elucidated by voluminous and rigorous experimentation. Thus although EBM correctly recognizes that basic science is an insufficient basis for determining the safety and effectiveness of a new medical treatment, it overlooks its necessary place in that exercise.
In part 1 I gave the first corollary of Baye theorem for poets, surgeons and the rest of us: extraordinary claims require extraordinary proof. Now I'll give you the second corollary: evidence based woo is still woo.
Monday, December 21, 2009
First a little background. If you've forgotten a lot of the biostatistics you once learned, as I have, the math behind Bayes theorem can be pretty stiff. But here are some resources for those who want to drill down:
Wikipedia article here.
Science Based Medicine posts by Kimball Atwood here, here and here.
Companion pieces in Annals of Internal Medicine here and here.
JACC article here.
To appreciate this in the proper depth read and understand these resources. (For me that'll mean several re-reads and referring to some background material).
For today's purposes, though, here's a primer on Bayes theorem for poets, surgeons and the rest of us. The non-quantitative aspects of the Bayes theorem are easy to understand because they conform to how clinicians think intuitively. Simply put, Bayesian analysis seeks to interpret new information in light of what was known before. A judge, for example, when sentencing an offender, takes into account prior offenses.
Let's start with the more familiar area of diagnostic testing and consider Bayesian analysis of ECG stress tests. Simple ECG stress testing is not commonly used today but it's a well studied and easy to understand example for our discussion. Bayesian analysis of treadmill stress test results was described in the classic paper by Diamond and Forrester. Suppose you choose 1.5mm ST segment depression as your threshold for a positive result. The specificity, determined from studies comparing treadmill testing to the gold standard of coronary angiography, has been estimated to be around 80%. Now here's where people trip up. It's common to confuse specificity with positive predictive value. The usual mis-perception is that specificity is the probability that the patient with a positive test has the disease. That's what positive predictive value is, but specificity tells us nothing of the sort. Specificity applies the assumption that the patient does not have disease and tells us the true negative rate. The specificity, which is solely a test characteristic, may be very high. The positive predictive value of a test with high specificity, on the other hand, may be high or low depending on patient characteristics that were known before (or independently of) the performance of the test (remember that Bayesian analysis always takes into account what was known before!).
Data from Diamond's article and other sources illustrate how patient characteristics influence positive predictive value (also often referred to as post test probability of disease) with a fixed specificity of 80%. For a 50 year old male, for example, with atypical chest pain (some features of angina but not classic) who has a positive test at 80% specificity the positive predictive value is around 90%. In other words he has a 90% probability of disease. A 35 year old woman with atypical chest pain and that same test result, on the other hand, would have a positive predictive value of 20%. In both cases the specificity of the test is 80%. (If those numbers seem low remember that these data were collected in the 1970's and preceded our current epidemic of obesity and type 2 diabetes. Application of this analysis today might require additional patient characteristics beyond age, gender and chest pain pattern. Nevertheless it suits the purposes of our discussion). Using this same model consider a 35 year old woman whose chest pain is right inframammary, localized to an area the size of a quarter and is pleuritic (non-anginal chest pain). Again assuming 80% specificity and that same positive test result, her positive predictive value (post test probability of disease) would be next to zero. For that patient a diagnosis of “chest pain due to coronary artery disease” would be an extraordinary claim indeed. It would take much stronger evidence than a positive treadmill test, even with a specificity of 80%, to diagnose CAD in that patient. In fact, nothing short of the gold standard of a positive coronary arteriogram would be convincing. That brings us to a corollary in our primer on Bayesian analysis for poets, surgeons and the rest of us: extraordinary claims require extraordinary proof!
In a subsequent post I'll illustrate how that little maxim applies to clinical research (hint: can you say woo?).
Friday, December 18, 2009
Thursday, December 17, 2009
What would be an ideal agent for treatment of acute heart failure syndromes? Such a drug should (1) improve hemodynamics without adversely affecting heart rate or blood pressure or increasing myocardial oxygen demand and without reducing coronary perfusion; (2) prevent further neurohormonal activation, favorably modulate the existing neurohormonal milieu, or both; (3) be applicable in the context of known evidence-based therapies, such as ACE inhibitors and β-blockers; (4) help control ventricular rate in atrial fibrillation; (5) have a formulation for intravenous use during the acute phase of heart failure and an oral formulation for long-term use; (6) importantly, improve symptoms and signs, decrease rehospitalization rate, improve survival, or all 3; and (7) be affordable for the millions of patients with heart failure throughout the world. This last point is particularly important because many patients with heart failure cannot afford newer, much more expensive drugs.
Digoxin has many of these properties.
Digoxin's much respected toxicity is an extension of known pharmacological effects and can be avoided by meticulous attention to dosing.
Wednesday, December 16, 2009
Conventional mechanical ventilators sense a patient effort by either a drop in airway pressure or a reversal in flow. The last and most slow reacting step in the chain of respiratory events is used to sense the patient effort. Hence, creating a system that is sensitive to hyperinflation, intrinsic PEEP and secondary triggering problems.
With NAVA, the electrical activity of the diaphragm (Edi) is captured, fed to the ventilator and used to assist the patient’s breathing.
Rather than sensing negative pressure or flow to trigger a breath the system senses neural traffic. This has the potential to markedly decrease work of breathing and fatigue. To avoid diaphragmatic disuse the level of support with each breath can be varied. The sensor is apparently an adaption of an NG tube. More systematic study will be necessary to determine its precise role.
In MRSA bacteremia vancomycin MICs of 1.5 or greater were associated with a 36.4% treatment failure rate.
This paper showed an atypically high failure rate in bacteremic patients regardless of the MIC, and a virtual guarantee of failure for MICs above 1.
For MRSA isolates with vancomycin MICs less than 0.5 µg/ml, vancomycin was 55.6% successful in the treatment of bacteremia whereas vancomycin was only 9.5% effective in cases in which vancomycin MICs for MRSA were 1 to 2 µg/ml.
This paper showed higher mortality in MRSA bacteremia treated with vancomycin when the MIC was above 1.
Tuesday, December 15, 2009
Up to 25% of patients admitted to the hospital with COPD exacerbation have acute PE. This prevalence equates to something between an intermediate and high risk Wells score even though by definition (“admission for COPD exacerbation”) such patients are lacking one of the two most heavily weighted Wells criteria (patient lacks a more likely cause of dyspnea). In related posts I noted that PE was the second most common cause of death in patients hospitalized with COPD and cited the meta-analysis referenced in the talk and one other paper.
10% of patients with unprovoked VTE will have a diagnosis of cancer within a year (paper here). The evidence, according to the speakers, does not support total body CT scans on all those folks, but thorough physical and standard laboratory examination as well as making sure standard screening recommendations are followed would be appropriate. When patients ask “what caused this blood clot” discussions often center on hereditary thrombophilia. How often do we advise patients of the increased cancer risk?
A five day antibiotic course may be all that's needed in most COPD exacerbations (paper here). Don't omit antibiotics, though, just because the patient doesn't have pneumonia, as there is evidence that antibiotics decrease mortality for AECOPD.
Dose vancomycin by actual body weight in obese patients, according to newly released IDSA guidelines for vancomycin use. (Earlier this year this consensus statement was released). Related post here.
Immunosuppressive therapy has no role, with the exception of catastrophic APS or a concurrent lupus flare as an indication for immunosuppression.
Higher than usual intensity anticoagulation (INR 3.1-4) is no better than standard intensity.
APS is believed to be a state of extreme hypercoagulability. It has been thus widely assumed that patients with APS who have had VTE should undergo lifelong anticoagulation. Unfortunately RCTs have not directly addressed the question. Because evidence from the general population supports indefinite anticoagulation for some patients with idiopathic VTE and patients with APS are at higher risk than the general population, indefinite anticoagulation is widely recommended for secondary VTE prevention in patients with APS.
Future studies may clarify the role of serologic profiling in patients with APS. Preliminary evidence suggests that, for example, patients with the lupus anticoagulant are at higher risk than patients who only exhibit cardiolipin antibodies. Patients who have cardiolipin antibodies, lupus anticoagulant and beta2GPI antibodies (“triple threat”) may have the highest risk.
In patients with VTE due to transient risk factors who are found to have laboratory evidence for APS there is little evidence to guide clinicians, and a finite duration of anticoagulant therapy may be suitable for some patients.
The authors' recommendations for patients with APS and arterial thrombosis differ little from those for the general population (antiplatelet therapy is as good as anticoagulation) although they tend to favor dual antiplatelet therapy over monotherapy. Nuances are discussed in the text.
Obstetric APS is discussed in some detail and not covered here.
Catastrophic APS is mentioned briefly and a detailed review is cited. Specific diagnostic criteria exist and the mainstays of treatment are prompt anticoagulant therapy, corticosteroids (if infection is not present), treatment or elimination of precipitating factors and treatment of an accompanying lupus flare if present.
Monday, December 14, 2009
In reality, handwriting issues aside, there is little wrong with "the old medical chart" from an information science perspective. It evolved over a century or longer to serve the needs of its users. It is a simple document in terms of organization, containing sometimes complex information but in an easy to find form (when maintained by humans properly) and in a presentation style that recognizes human cognitive limitations in very busy, complex social environments such as patient care settings.
Its quasi-duplication in electronic form would serve medicine well.
...In fact, politically speaking, health IT can be viewed as a cross-occupational invasion of healthcare by the IT industry. (Other invaders are at work also, but I am only considering the IT industry here.)
The latter industry is largely healthcare-dyscompetent or incompetent .
What right do the domain-dyscompetent occupants have to tell the occupees, the latter rigrorously trained in clinical medicine through years of both classroom and grueling practical experience, and in the record keeping paradigms developed over centuries, how to maintain their records and perform their processes?
What arrogance is it that drives the the occupants to tell the occupees to stop complaining about the terms of the occupation - seriously deficient experimental health IT applications - and get in line with the methodologies and preferences of the occupants?
Some will view this as Luddite mentality. I view it as healthy skepticism.
When a family member in a patient's room suspects their loved one's condition is rapidly worsening or a serious new health concern is developing, their instincts are often correct.
A new system at Vanderbilt gives families a powerful tool to quickly call for help.
The Family Initiated Rapid Response Team program is a lot like calling 911, but when a family member dials 11111 from a Vanderbilt patient's room, a LifeFlight operator dispatches the rapid response team.
Via the VUMC Reporter.
Friday, December 11, 2009
However, the resultant disruption and restriction of physicians' sleep produce demonstrable neurobehavioral impairments that may threaten other fundamental professional mandates, such as that of primum non nocere (“first, do no harm”).
Thursday, December 10, 2009
The ABFM is now also offering a hospital medicine path for maintenance of certification, so that now both internists and family practitioners can earn recognition of focused practice in HM. This may accelerate the entry of family practitioners into hospital medicine. It may help address the current shortage of hospitalists and will likely change the landscape of hospital medicine in a number of interesting ways.
The first secure exam will take place October 25, 2010.
The self-evaluation module in the hospital medicine pathway is “not going to differ appreciably [from those used in the existing MOC process], but the vision going forward is [to have] more hospital-focused sets, on things like patient safety and transition of care,” said Dr. Wiese, who is a professor of medicine at Tulane University in New Orleans.
The practice improvement module for hospitalist certification “will focus on the ability to interact well with a team,” he said. “The candidate would sign up, would identify some practice area, collect data on the practice, design an intervention, and collect data again after a few months to show an improvement in practice delivery.”
The secure exam will differ from the standard internal medicine exam “in that it has much greater hospital medicine–focused content, but also questions on quality, patient safety, and transitions,” he continued.
Wednesday, December 09, 2009
According to a paper in the Journal of Thrombosis and Haemostasis the timing of the thrombocytopenia is highly predictive:
The high frequency of thrombocytopenia in post-cardiac surgery patients makes it challenging to diagnose heparin-induced thrombocytopenia (HIT). Two platelet count profiles are reported as indicating possible HIT in these patients: profile-1 describes a platelet count fall that begins between postoperative days 5 to 10, whereas profile-2 denotes early-onset thrombocytopenia that persists beyond day 5.
...Thrombocytopenia that begins between 5 to 10 days post-cardiac surgery is highly predictive for HIT. In contrast, early-onset and persisting thrombocytopenia is usually caused by non-HIT factors with coinciding heparin-dependent antibody seroconversion.
Many patients with antibodies coincidentally have non-HIT related thrombocytopenia.