« December 2008 | Main | February 2009 »
January 29, 2009
JUPITER
As mentioned in the previous post, JUPITER is the latest of the GALAXY trials to be announced. It is particularly welcome to the commercial interests which sponsored it because it extends the use of this drug to 20% of the population that otherwise could not have been convinced to spend the money.
As also mentioned in the previous post, I have avoided spending any precious time on the Statin studies, in large part because the basic premise and justification, upon which all these subsequent studies have built, is fatally flawed. There is perhaps a mild statistical significance in the decreased measures of heart or brain disease. But there is no meaningful improvement in these, and no improvement at all in overall mortality.
I will undoubtedly be challenged on that last statement. But I would refer you to Worried Sick or to Joel Kaufmann's articles and letters in JAPS especially this one.
Back to Jupiter. Basically, low risk patients older than 50 for men or 60 for women (why the difference? because that is the only way to make the results statistically significant) were screened for low cholesterol but elevated hsCRP (high-sensitivity C-reactive protein). Those qualifying (only 20% of those screened) were enrolled and put on very high doses of The Statin. (They knew from other studies with other statins that low doses would not show an effect.) These news items (I realize some of these links require a password, but you can't have opened your eyes in the last two months without seeing something about Jupiter) give a good idea of the significance of the study, which has been touted as "paradigm-changing" and "one of the most important clinical trials in the long history of statin studies."
But it rather is just another attempt in the long train of confused commercial/clinical jumbles. I do not fault the drug company in question for trying to create a new market for their very expensive drug. But when this is confused with good medicine, there are problems. The very minimal absolute risk reductions for so-called hard events (death, heart attack, and stroke) require treatment of 170 patients to prevent one yearly event.
At more than $100 per month, medication costs alone would exceed $300,000 per event per year (more than $500,000 to prevent a death). It is not that a price can be put on life. Rather, it is absurd to spend so much money for so little benefit. If you are spending your own money, you answer for that. If you are spending mine, quit.
Posted by Robert Maddox at 12:46 PM | Comments (0)
GALAXY
I admit that I have been negligent in keeping up with all the medical literature. (I will address that issue soon). But until I was looking at the JUPITER trial (more later), and was laughing at the names of related trials, I did not realize what I have been missing. Back in 2002, it was announced that there would be a series of trials "under the umbrella name of the GALAXY program, which aims to enroll over 80,000 subjects from more than 30 countries to test a range of effects of statin therapy." This is the fun side of hubris:
Glossary: Expansion of Selected Trial Acronyms
CELESTIAL - Changes in ELEctrocardiographic Signs of Myocardial Ischemia: a TrIAL with rosuvastatin
COMETS - A Comparative study with rosuvastatin in subjects with METabolic Syndrome
DISCOVERY - Direct Statin COmparison of LDL-C Values: an Evaluation of Rosuvastatin therapY
LUNAR - Limiting UNdertreatment of lipids in ACS with Rosuvastatin
MERCURY I - Measuring Effective Reductions in Cholesterol Using Rosuvastatin therapY
MERCURY II - Measuring Effective Reductions in Cholesterol Using Rosuvastatin therapY
ORBITAL - Open label primary care study: Rosuvastatin Based compliance Initiatives linked To Achievement of LDL goals
STELLAR - Statin Therapies for Elevated Lipid Levels compared Across doses to Rosuvastatin
Notice that all the studies concern just one of the statins. The article also mentions other trials of different statins with less exciting names:
ENHANCE - Ezetimibe and Simvastatin in Hypercholesterolemia Enhances Atherosclerosis Regression
SEAS - Simvastatin and Ezetimibe in Aortic Stenosis
Posted by Robert Maddox at 12:29 PM | Comments (0)
Worried Sick Revisited
After writing the positive review of Worried Sick a few weeks ago, I had intended to use another post covering some of the negatives. Instead, I will just post this link to a great review by my mentor (who deserves no opprobrium for that circumstance). His insightful analysis of Hadler's presuppositions, and the conclusions that result, are important reading for any one evaluating the scientific method and its fruit.
Posted by Robert Maddox at 9:41 AM | Comments (0)
January 28, 2009
What to do when the experts disagree?
As mentioned previously, the issue of the crud is copious, meaning that I am seeing many snotty nosed people. Rather than just assume that I am right, and doing the best for these suffering people, I have attempted to review the available evidence. I recalled an article from the JFP, and a more recent Clinical Guidelines from FPN. The latter follows the ENT guidelines from the American Academy of Otolaryngology--Head and Neck Surgery. The former is a PURL (Priority Updates from the Research Literature from the Family Physicians Inquiries Network). The PURL editor is the fellow under whom I was doing my research project for FP during med school. I regret that I did not learn more from him.
The FPN Guidelines lay out some nice terminology distinguishing acuity (acute, subacute, chronic or recurrent acute) and source (bacterial or viral). The only point of value here is that 4 weeks is considered a normal duration for rhinosinusitis. Even up to 12 weeks doesn't earn you chronic status.
The big question here is whether antibiotics are helpful in purulent rhinosinusitis. It turns out that they are -- with an NNT of 15. Fifteen people have to be treated with antibiotics for one to be cured at 8 to 15 days after initiation of treatment. That means that 14 of those people receive no benefit from the time and expense of taking the antibiotics. Based on the risk of side effects, this PURL concludes that there is no indication for antibiotics in routine purulent rhinosinusitis. Which means that antibiotics are not helpful. (They do make an exception of course for those that are seriously ill.) However, there was a meta-analysis done the year before that shows an NNT of 7. The studies it was based on had advanced imaging and cultures to help distinguish viral from bacterial. Furthermore, the FPN Guidelines are based on the expert opinion, and no hard evidence, when recommending antibiotics if the patient doesn't improve in 7 days. The subsequent study by Young which the PURL evaluates trumps the expert opinion.
Continue reading "What to do when the experts disagree?"
Posted by Robert Maddox at 9:14 PM | Comments (0)
January 26, 2009
EHR stimulation
Mark Horne has blogged about the EHR provision in the economic stimulus bill. I have had neither the time nor inclination to worry about the new administration's shenanigans. I am too busy seeing the sick and the miserable to use the EHR right now. I know that others tout the wonders of them. And I do see a few advantages.
The sharing of health information bothers me most because of the commercial exploitation potential. The way the internet entrepreneurs currently exploit every link we take is irritating. It will be downright frustrating, if not intrusive, for the commercial exploiters to know all your health concerns.
And what of potential employers and insurers (assuming there is any of that left after the new administration is done centralizing everything)? Gattaca will look mild in its oppression.
Posted by Robert Maddox at 9:52 AM | Comments (0)
January 22, 2009
pneumonia vaccine questioned
It is somewhat gratifying when a study supports my perspective. This meta-analysis on the efficacy of pneumococcal vaccine demonstrates minimal benefit of the vaccine in preventing pneumonia, even pneumococcal pneumonia, much less death, especially in the aged and ill, which is the population that needs it the most.
The WHO is not convinced. There are other studies that demonstrate some benefit. And when the goal is complete freedom from risk, any benefit is worth any cost.
Posted by Robert Maddox at 10:02 AM | Comments (0)
January 19, 2009
SNOT
Today's winner for a great name, without hubris, is SNOT-20 (Sino-Nasal Outcome Test), a 20-item validated, self-administered survey.
This important piece of information was brought to my attention in an article discussing the use of nasal irrigation in patients with chronic rhinosinusitis. I, like everyone else in primary care this time of year (at least in the South) am seeing many patients with acute and chronic rhinosinusitis. I recommend sinus irrigation to every one of them. In the past, it was homemade snorted from a glass. Admittedly, that was very difficult for most to work up the nerves to try. Then came the WaterPik, but that was expensive. Then someone got rich off the little Aladdin's lamp, which has a tradename. Now there are very inexpensive squeeze bottles, with pre-measured packets.
The study explored in this PURL (Priority Updates from the Research Literature) was limited to Chronic R-S. It excluded children, those with recent sinus surgery, a respiratory infection within the past two weeks, and the use of nasal saline within the past month (which means those that already knew it worked cannot participate). So this does not apply to acute conditions, or to many of those with chronic conditions. They concluded that, using the SNOT criteria, irrigation was better than nasal spray. Unfortunately, they did not compare to placebo (how could one?). And of course, it could not be blinded. But at 8 weeks, there was an absolute risk reduction of 21% with irrigation over spray (of having frequent sinus symptoms). Using SNOT-20, there was a 7-9 point reduction.
Now look at the SNOT-20 link. One can reduce one's SNOT-20 score by 10 points by changing from "problem as bad as it can be" to "severe problem" in just half of the categories. I am just not sure how significant that really is. As much I encourage this practice, this is just not that meaningful. So for the record, even my limited hubris is too much.
Posted by Robert Maddox at 4:47 PM | Comments (0)
January 14, 2009
English v Germans again (on Afib cardioversion)
OK, admittedly, the country of origin probably has nothing to do with the results. I am referring to the German study on electrocardioversion of atrial fibrillation and the English opinion on the same problem. Admittedly, NICE (the English agency that makes pronouncements on efficacy - not to be confused with the NICE in That Hideous Strength since they are so similar as to be easily confused) apparently recommends electroversion as an option. Having worked the ER for decades, I do not know what the common approach is to chronic afib in the US. We occasionally have to electrovert a patient in the ER that does not respond to chemical arrythmics or is in distress.
The English review relies on AFFIRM (Atrial Fibrillation Follow-up Investigation of Rhythm Management) which not only showed that cardioversion was associated with higher mortality than rate control, but also showed that the anti arrhythmic meds were also associated with higher death rates and more frequent hospitalization and arrhythmias. (We have always know that the worst side-effect of anti-arrhythmics is worse arrhythmias, just as the anti-psychotics can cause psychosis.)
It is no surprise that many darling medical interventions are found to be relatively useless or even dangerous.
What is curious is this statement: "Paradoxically this evidence has derived from trials designed to prove the effectiveness of the procedure." It is the hubris of medicine to think that the study must always confirm our hypothesis. We do studies to support a particular approach and are surprised (or confused) when it does not. Change the parameters until it does.
Posted by Robert Maddox at 1:42 PM | Comments (0)
Con-us
There is a study that competes for first place in my CON-US (COmmitting hubris in Naming Useless Studies) category: GRACE (Global Registry of Acute Coronary Events). Among other things, GRACE has produced a risk stratification score for chest pain (as has the aptly named PURSUIT - platelet glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy and TIMI - Thrombolysis in Myocardial Infarction).
[I am reading Shifting the Paradigm for ACS (Acute Coronary Syndrome) Management. I hope to make some comments about it soon.]
Grace: God's Riches At Christ's Expense. Or more accurately, the relationship of favor by which God calls us kin, and exhibits to us his kindness.
What are these people thinking?
Posted by Robert Maddox at 12:00 PM | Comments (0)
Making fun with statistics
Many people, including most doctors, don't understand statistics (think Mark Twain quote here). But I ran across this funny explanation of lead-time bias that might be helpful to some.
Posted by Robert Maddox at 8:24 AM | Comments (0)
January 13, 2009
Nutty changing medicine
This is old news but it is relevant to friends and family, and a great illustration of my principal principle: there is no truth in medicine.
A decent study on nuts and popcorn in patients with diverticulosis not only showed no increased harm from ingestion of the long-forbidden foods, but actually showed an "inverse association between the consumption of nuts and popcorn and the risk of diverticulitis." And no association with bleeding.
I learned in school and residency that the average half-life for medical dicta is about 15 years. Interestingly, this was a cohort study of 18 years' duration.
Posted by Robert Maddox at 5:11 PM | Comments (0)
January 8, 2009
new wine
The recent pastors conference included a discussion of Christ's directives about new wine in old wineskins. An oft-missed point in these discussions is that wine matures and "gets old" and wineskins age. And that is the way God intends it. So Jesus is telling us to be aware that he works in this way and don't be too confused about it.
The church is the seed form, or the alpha form, of the other institutions in society. So as God brings new things into the word (like techie things), we shouldn't try to constrain them in the old wineskins. Now it is true that the old wine is better, but only for the time being. The new wine becomes old eventually. And it is to be consumed.
I am struggling with how this applies to medicine. I know (and trust God) that the changes that are being brought about in medicine are in his plan. I know also that they will look bad, and be brought about by wicked men with evil motives. And when he suddenly and amazingly demonstrates that what was intended for evil (or self-aggrandizing, or confused motives, or greed) was actually for good, as we are told clearly was the case with Joseph, and more clearly yet in Christ's death.
Posted by Robert Maddox at 4:43 PM | Comments (0)
EMR
An article in American Medical News indicates that EMR's (electronic medical records) are being expected and demanded by graduating doctors, since they used them in training, and since they are "tech-savvy."
The medical organization that is fighting for good medicine (AAPS) has generally been against the EMR push for a variety of reasons, particularly privacy concerns.
I can neither fight nor applaud this development. I have tried to use an EMR in the office, and found it fun, but time-consuming. I am sensitive to the privacy issues that it raises, but doubt that will be a huge problem. I do not think it will improve patient care much. The drug interaction software built into it has been a slight help. But it mostly improves documentation for coding purposes. It will not prevent medical errors. Like the T-system popular in the ER, it will help the less alert, or less trained, remember to ask or look for certain problems. But I suspect that there is more documented on a check system than is actually examined on the patient. And if you need an EMR to make you think the patient is having a stroke or heart attack, they probably will survive it anyhow, or will die under your care regardless.
Posted by Robert Maddox at 4:05 PM | Comments (0)