Antibiotics

November 25, 2011 9:33 AM

The US Center for Disease Dynamics, Economics and Policy runs a non-profit, Extending the Cure, which conducts policy research to extend antibiotic effectiveness, and ResistanceMap, which generates interactive analysis tools and maps regarding antibiotic use in Europe and the US. The most recent ResistanceMap visualizations indicate that the US Southeast overprescribes antibiotics at a high rate compared with the rest of the country. Science journalist / "Superbug" blogger Maryn McKenna speculates (while acknowledging that correlation ≠ causation,) that the map might also indicate a link between overuse of antibiotics, obesity, diabetes and stroke.

A recent commentary published in Nature raised additional/related concerns about antibiotic overuse. More.

posted by zarq (30 comments total)

Or, how about, the South is poorer on average, thus more people go to the doctor in order to avoid more time off from work that they can't afford. Most hourly jobs don't give sick time, and missing work means being short money or worse fired
All of those other conditions are symptoms of poverty as well.
posted by strixus at 9:46 AM on November 25, 2011 [2 favorites]

The excessive use of antibiotics may also be playing a role in the increase in allergies seen among young people.
posted by Blasdelb at 9:46 AM on November 25, 2011

There is already a well-characterized interaction and cluster among obesity, heart disease, and diabetes. These are unhealthy individuals. It doesn't seems surprising that these same people would end up taking more antibiotics for what ails them. I find her speculation about a link between antibiotics and these conditions uncompelling without at least a theory about a mechanism.
posted by Mercaptan at 9:48 AM on November 25, 2011 [1 favorite]

RTFA dammit.
posted by lalochezia at 10:00 AM on November 25, 2011 [2 favorites]

The error here isn't confounding but rather the ecological fallacy.
posted by docgonzo at 10:02 AM on November 25, 2011

the South is poorer on average, thus more people go to the doctor

How does this follow? Poor people are less likely to have health insurance, especially if they have jobs with no paid sick time.
posted by desjardins at 10:09 AM on November 25, 2011

I'm a little tired of people waving around "correlation is not causation" incorrectly.

The correct phrase is "correlation is not direct causation". There are time when correlation is coincidental, and where correlation is directly causational, but there are more times where the two events do have a common cause.

Forex, strixus' posit about all of these cases being increased by poverty. I'm not asserting that this is the case, but it's an example of how dismissing the correlation could well make you miss the cause.

What correlation really means is "dig harder".
posted by eriko at 10:10 AM on November 25, 2011 [9 favorites]

As Mercaptan points out... if these are simply the areas with a higher rate of health problems, doesn't it stand to reason that they're going to be consuming every medical supply in greater quantities? The author is using antibiotic prescriptions per person rather than something like antibiotic prescriptions per hospitalization or per doctor's visit.

She also points out that some of these areas have poorer access to healthcare than others, which says to me that people are going to go longer between doctor's appointments and hospital visits and thus need more extreme treatment on the occasions when they do get care.
posted by XMLicious at 10:10 AM on November 25, 2011 [1 favorite]

I remember discussing questions of antibiotic resistance with a doctor in my FB circle a while back. I have a history of having colds turn into sinus infections and thence into bronchitis, and because of that history and other health issues, my doctor is pretty quick to prescribe antibiotics to me if I get a cold and it looks like it might turn nasty. The doctor's comment was that overprescribing antibiotics to humans was a significantly smaller problem in terms of antibiotic resistance than the use of antibiotics in feed for meat and dairy animals.

If antibiotics are really causing the sorts of health problems the research in the Nature and NYT links suggests, I have to wonder what the effects of antibiotics in feed for those of us who consume meat/dairy/etc.
posted by immlass at 10:11 AM on November 25, 2011

I read TFA, this: "That lines up with the work of Martin Blaser of New York University, who has proposed that taking antibiotics permanently kills off beneficial bacteria in the gut — and might therefore be responsible for changes in nutrient absorption and for the rise in obesity and diabetes." is not a working theory. What kinds of changes in nutrient absorption generated by gut flora alteration might lead to obesity and diabetes are we talking about? Until then, it's so general that it's untestable.
posted by Mercaptan at 10:12 AM on November 25, 2011

Mercaptan, the potential mechanisms for a causative link between decimating the gut flora, a vital organ involved in most of the metabiolic heavy lifting our livers don't have the tools for as well as regulating nutrient adsorption and satiety, and obesity should be obvious.
posted by Blasdelb at 10:12 AM on November 25, 2011 [2 favorites]

This paper from PNAS a while ago went into some of the precise mechanisms suggested by this paper. The authors from the FPP didn't go into precisely how, but their work is supported by folks who have for similar questions.
posted by Blasdelb at 10:21 AM on November 25, 2011

The ResistanceMaps should really take farm animal dosing into account. Hog and chicken farm numbers might form a similar correlation with resistance for the Southeast..

IMHO, Antibiotic use in humans is a big red herring for the evolution of resistance. Routine use of antibiotics in animal feed needs to be outlawed. You couldn't devise a better way to breed resistant bacteria if you tried. Losing all of our antibiotics is not worth $2.99 chicken breasts.

We've been feeding drug members of all these classes to farm animals continuously (not just a one or two week regimen) for years. Although the human version of an antibiotic may not be available for farm use, a close relative is usually available that works through the same mechanism. When resistance inevitably arises, the same genes usually inactivate all the related antibiotics in the same chemical class.

This is not to say overuse in humans is OK, but if I want to put money towards prventing resistance, something like the Preservation of Antibiotics for Medical Treatment Act would go much farther for the effort (except factory farming corporations have and will lobby against it).
posted by benzenedream at 10:49 AM on November 25, 2011 [6 favorites]

Blasdelb: Thanks. Actually this Nature article is more compelling, where they inoculated flora-free mice with gut flora from leptin-defective obese mice and were able to measure a significant body fat increase (other then that they obtained the flora from Lep-defective mice, rather than lifestyle-obese mice) as compared to flora-free mice inoculated with normal-type flora.

Except I don't want to know how humans are inoculating each others gut flora.
posted by Mercaptan at 11:03 AM on November 25, 2011

I'm a little tired of people waving around "correlation is not causation" incorrectly.

The correct phrase is "correlation is not direct causation". There are time when correlation is coincidental, and where correlation is directly causational, but there are more times where the two events do have a common cause.

It's more like, correlation doesn't tell you the direction of causation. If A and B are correlated, it doesn't mean that A causes B, but it does mean that there is some causal relationship, possibly involving a hidden variable causing it. C coincidental correlation would only show up if you have a data set small enough to cause the error.
posted by delmoi at 11:05 AM on November 25, 2011

It's more like, correlation doesn't tell you the direction of causation.

Even more to the point: as Hume pointed out a couple of centuries ago, the only thing we know about causation is that it represents a constant and repeatable correlation. The saying really ought to be "correlation is not always causation."
posted by yoink at 11:09 AM on November 25, 2011 [1 favorite]

It's more like, correlation doesn't tell you the direction of causation. If A and B are correlated, it doesn't mean that A causes B, but it does mean that there is some causal relationship, possibly involving a hidden variable causing it. C coincidental correlation would only show up if you have a data set small enough to cause the error.

I think the original use of the phrase was in reference to Congress, which usually refuses to regulate a well used, profitable chemical (particularly synthetic organics used in agriculture) until the mechanism of toxicity is discovered. Just showing that 99% of people (in the real world) and animals (in carefully controlled lab experiments) exposed to it develop three breasts usually isn't enough.
posted by Slackermagee at 11:11 AM on November 25, 2011

I have a history of having colds turn into sinus infections and thence into bronchitis, and because of that history and other health issues, my doctor is pretty quick to prescribe antibiotics to me if I get a cold and it looks like it might turn nasty.

As an MD reluctant to hand out antibiotic prescriptions (who often gets crap for it from patients, btw), I'd like to point out that sinusitis and especially bronchitis still tend to be viral infections. A nasty cold progressing to one of these does not automatically warrant antibiotics, despite what many patients believe, and despite how many of my colleagues act.

Unfortunately, a patient who has gotten antibiotics once for bronchitis will forever think that they need more of the same if bronchitis strikes again and will look askance at any doc who tries to convince them otherwise. One of my biggest professional pet peeves...
posted by thelaze at 11:14 AM on November 25, 2011 [5 favorites]

A nasty cold progressing to one of these does not automatically warrant antibiotics

That may be true for most patients, but since you're not familiar with my medical history, I'd appreciate your not using it to score points about your pet peeve. No offense intended, but people who aren't familiar with my health history telling me how to manage my health and/or implying I'm a drug seeker is one of my personal pet peeves.

In any case, if people are really worried about the effect of antibiotics on the population, antibiotics in feed are still a bigger issue than how doctors medicate the colds of people who actually go to the doctor's office for them (not everybody, not even always me).
posted by immlass at 11:35 AM on November 25, 2011

I'm not sure where in my comment I gave anyone a directive on how to manage their health. I'm usually fairly prudent to only dispense medical advise to patients directly under my care.

My comment was based on the countless times I've encountered patients demanding antibiotics for viral infections who were given the impression that antibiotics were warranted by their past encounters with the medical profession.

It would serve the public well to know that respiratory tract infection are typically viral. Certainly there are situations where antibiotics are important (ie, COPD), but in the vast majority of cases they are not, and there is a real lack of awareness of this fact that my colleagues are partly to blame for.

I'm not scoring points. I consider this an important and relevant health issue often neglected.
posted by thelaze at 11:46 AM on November 25, 2011 [10 favorites]

It's more like, correlation doesn't tell you the direction of causation.

This statement might be true in cross-sectional studies but breaks down when longitudinal or survival models are considered.

It has always been my belief that anyone pulling out the correlation/causation objection is demonstrating they don't understand Bradford-Hill well enough.
posted by docgonzo at 11:58 AM on November 25, 2011 [1 favorite]

As criminal as the use of many antibiotics in feed is, to say that overuse in humans is a red herring goes way to far. Just look at Staphylococcus aureus, the original bug antibiotics were meant to fight against.

Penicillin resistant S. aureus was isolated just a couple years after the introduction of penicillin, and way before its use in factory farming. Since then, no antibiotic has had 100% efficacy for more than a few years. Yet, there really is no meaningfully plausible etiological connection between agricultural use of antibiotics and resistance in S. aureus (with the possible exception of the rare zoonotic cases in dairy workers) as it is an obligate commensal with no environmental part of a life cycle.* We know pretty well how resistance evolved in S. aureus; it started in hospitals and, at least as of a few years ago, almost all of the community associated outbreaks were traceable directly to hospital associated strains. Muti-drug resistant S. aureus now kills more people in the United States than AIDS.

Of course it is way to plausible that agricultural use of antibiotics is contributing to some of the epidemics, and the use of medically relevant antibiotics should absolutely be banned for that reason, but to say that is is a major, much less the primary, cause next to human use stretches the imagination too far.

*If someone has agricultural products or waste up their nose, where S. aureus lives, I would bet they have bigger problems to deal with than antibiotic resistance.
posted by Blasdelb at 12:00 PM on November 25, 2011 [5 favorites]

That may be true for most patients, but since you're not familiar with my medical history, I'd appreciate your not using it to score points about your pet peeve. No offense intended, but people who aren't familiar with my health history telling me how to manage my health and/or implying I'm a drug seeker is one of my personal pet peeves.

In any case, if people are really worried about the effect of antibiotics on the population, antibiotics in feed are still a bigger issue than how doctors medicate the colds of people who actually go to the doctor's office for them (not everybody, not even always me).

Hate to break it to you, but a nasty cold does not care about your medical history or the number/dosage/strength of the antibiotics you are taking. That nasty cold is very much un-living. It is a product of life which does not live itself. It is abiotic. Something which kills living cells (an antibiotic) is about as useful against the cold (rhinovirus) as a strong salt solution is against a halophile. Well, wait, that last one may have been a little geeky... uh... can I phone a friend on this analogy?

South Park: "How do you kill... that which has... no life...?"

While antibiotics in feed are a problem, I'm more worried about all the nasty little buggers that are human specific (and very much alive) which are getting more and more resistant to antibiotics. Usually these critters are benign but God help you if they evade your body's immune system beneath the skin.

A far more pressing issue would be Tenacious B. And by that I mean XDR TB, extremely drug resistant Tuberculosis. From the papers I've read and the presentations I've listened to, XDR TB owes its existence to people not finishing their goddamn course of anti-biotics. Blame lazy people, blame big pharma's crazy prices, blame whoever you want. Its a scary fucking disease now and we'll be seeing the return of sanitariums if this thing ever leaves those two small regions in Africa.
posted by Slackermagee at 3:40 PM on November 25, 2011 [2 favorites]

Slackermagee, It already has...

Though, as this becomes a bigger deal, which it will, we do still have an advantage that should be able to keep the epidemic under some measure of control. We have quite wisely steadfastly refused to give childhood vaccinations against TB, which are only partially effective and only for a short time, and are thus still able to test for TB allowing for quick and effective quarantine. Then again, if you think we've shot ourselves in the foot with the anti-vaxxers, just wait until little old granny doesn't want to enter quarantine and FOX news doesn't think she should have to. We're headed for the kinds of necessary public health decisions our great grandparents were tough enough to make, but I doubt we are.
posted by Blasdelb at 3:53 PM on November 25, 2011 [1 favorite]

Except I don't want to know how humans are inoculating each others gut flora.

I believe the working theory is that you get your first set on the way out of your mother. But babies seem to interact with the world by putting it in their mouths, so it's not like they'll miss out if their mothers are uncharacteristically clean. If yours ever go missing try licking a bunch of random surfaces. You'll have a pretty diverse population in no time. (I don't seriously advocate this, but hey, it's worked for me.)

That's kind of the hole in this theory - it's not like your stomach is some sort of level 4 bio-containment facility and a lot of engineering went into setting you up with some perfect set of commensal bacteria. For antibiotics to be causing the problem, everyone would have to be born of a mother with some kind of magical perfect set of flora, which nothing on the outside could ever hope to compete with, but then you take antibiotics for strep throat, ruin the whole thing and the bad evil intestinal flora gets in, completely takes over and then you have a stroke. Or, you could go kiss your mother on the cheek or something after your last dose of antibiotics, inoculate yourself with a culture of perfectly balanced motherly flora and, with a doubling time of a half hour you only need to pick up one solitary bug and you'll be back to your nominal 1,000,000,000,000,000 bacterial cells by the next day.
posted by Kid Charlemagne at 4:11 PM on November 25, 2011

That may be true for most patients, but since you're not familiar with my medical history, I'd appreciate your not using it to score points about your pet peeve. No offense intended, but people who aren't familiar with my health history telling me how to manage my health and/or implying I'm a drug seeker is one of my personal pet peeves.

If you don't intend offense, don't be so damn quick to take it. thelaze is correct. The vast majority of sinus infections and bronchial infections are viral and not responsive to antibacterial drugs. You may not fit that pattern (although I doubt it unless you are prone to secondary bacterial infections pursuant to the viral infections that cause the vast majority of upper respiratory acute infections), but there is absolutely no question that doctors over-prescribe antibiotics to satisfy the demands of patients who are ignorant of the science involved.
posted by spitbull at 4:17 PM on November 25, 2011

This is a really nice post zarq--previous commentators have found lots of interesting ways to enter the discussion, as well. I've been sort of looking for ResistanceMap or US Center for Disease Resistance to enter the blue, especially during cold and flu season. Also, it's been awhile since I've read McKenna (stupid grad school), and so the link taking a look at possible correlations that ResistanceMap may indicate is interesting.

My first impulse is to educate about population and sample selection because I think that if more folks understood just those two statistical concepts than nearly 85% of trash healthcare reporting by the media would stop, already, but I can't resist entering into this from the perspective of someone who is on the front lines of the resistance issue--I'm in pediatrics (which is basically infectious disease in action) and in a few short months will have my DEA# and be prescribing--as it is I prescribe under my attending physician's license, and so have been and am already making decisions about things like antimicrobials and when they are appropriate for kids. Kids--who are actively developing the immunity they will have for the rest of their lives under my care.

A couple of very low-level educational issues come to mind that thelaze has already pointed out--one, basic immunological literacy, even among well-educated patients, is extremely low. Two, clinicians do not invest in providing patients with accurate, evidence-based, and effective supportive care treatment plans for viral illness that would offset dissatisfaction with a clinician's "failure" to prescribe antimicrobials. Right now, what I see out in the wilds, is that patients are hungry for actual healthcare. Often the only care they receive is a Rx. If they don't get one, it's a wasted copay (or income) because no care was provided (even if clinical judgement was). I see a patient asking for a Rx as a patient asking for care. So I read a lot in the literature about evidence-based ambulatory care for viral illness. I don't rely on what gets passed down from other clinicians as habitual practice (like the BRAT diet for gastro, which is NOT evidence based), but create treatment plans for patients based on what meta-analysis tells me. A lot of good clinicians compile these kinds of treatment plans into reproducible "prescriptions" to hand to patients and explain. There's actually a whole world beyond "fluids and rest" that can empower patients.

Immunological illiteracy is a lot harder to tackle. I'm fortunate to be in pediatrics because kids are very responsive to expert education. I want kids to grow up to be questioning and empowered consumers of healthcare, so when I can, I talk to even really little kids about how vaccines work and the differences between viruses and bacteria and the different ways they make somebody sick. Sometimes, even their caregivers listen. Why the hey-ho this kind of basic human biology isn't taught in school to the most primary of grades, I'll never be able to understand. Kids, until middle childhood, have 8-10 viral illnesses a year. Which means they are nearly always either in the prodromal or active stage of a viral illness for, basically, ten years straight. How cruel is it that we never explain any of this to them, but instead, drag them, wordlessly and with much frustration, to the provider's office, complaining (right in front of them) that they are sick AGAIN and what is WRONG with them? Then, we don't leave until somebody gives us a Rx for "the pink stuff," AND THEN we come back 48 hours later demanding to know why this kid has the nerve to have DIARRHEA on top of everything else and what can we script them? And where is their ENT referral? Because surely, it's NOT NORMAL to get this many sore throats?

But I don't blame patients or families. It only drives me crazy at the systemic level. Multiple for-profit pharmacies offer free antibiotics in multiple treatment classes, in-store minute clinics are having some kind of freaking love affair with azithromycin this winter, schools and daycares are dictating medical treatment as much as insurance companies by not allowing kids go to school unless they have "24 hours of antibiotics" down their gullets or squirted into their eyeballs (even though 85% of conjunctivitis is *guess what* virally mediated) and health education is triaged somewhere below a budget for the underwater chess club. And parents want their kids to feel better. They want to care for them. They bring their feverish, rashy, snotty child into clinic and want somebody to do something. Plus, there are whole cabals of other parents ready to judge both your parenting and your provider's standard of care--it's not even enough to tell someone their provider believes your kid will be fine if their provider gave their kid a rocephin shot for "the same thing."

Commensal flora, much of it yet to be really understood or discovered, has evolved to work closely with our innate, humorally and cell-mediated immune system and is comprised of a fascinatingly complicated network of bacteria, viruses, parasites, and (as I've learned from Blasdelb) phages. Chemical therapy is a shockingly clunky and unsophisticated defense against disease, in comparison (did you know, for example, that there is no evidence basis for the number of days to Rx an antibiotic? We literally have no idea how long a patient needs to take one for it to be effective against a bacterial colony, and too little kill and/or overkill is precisely what has landed us in this hot resistance mess in the first place). It doesn't surprise me at all, then, that resistance and other immunologically mediated diseases like diabetes may be correlated.

And the biggest secret? Kids are supposed to get sick. A lot. Common viruses, like enterovirus, have dozens of serotypes, and the more of them, every year, a kid introduces to their system, the smarter their immune system for the rest of their lives. It doesn't mean you leave your snotty-nosed baby in a corner, we care for our young; we work hard to comfort these ridiculously immature human young. And our big prefrontal cortexes that take so long to mature have worked out that there are better ways to expose kids to the more virulent and dangerous organisms, so we torture them with needles so they won't be tortured with morbid disease. Until we stop thinking that we can somehow outsmart and overcome much of what our bodies have already figured out, we're going to kill ourselves with the pink stuff. We have so much to learn and antibiotics were never meant to be the final answer--just one of them (because, don't get me wrong, well-ordered antibiotics are screamingly important in both ambulatory and in-patient care). But we've lost research focus in other ways to support and augment immunity--it makes me weep to think of everything that we've learned about immunity from the last 20 years of HIV research that we haven't even begun to touch in our search for the next shiny-new macrolide.

Also, WASH YOUR HANDS.
posted by rumposinc at 7:45 PM on November 25, 2011 [10 favorites]

Incidentally those phages rumposinc mentioned make fantastically non-toxic, strain specific, and self replicating, if challenging, antimicrobial agents.

Resistance would still be a problem, but would suddenly be an addressable one as there will always be another effective phage to isolate as a replacement. In the Republic of Georgia where they have been producing therapeutic phage cocktails for the Soviet (and now former Soviet) Union since the thirties, they have been generating a new cocktail every six months. They are also attractive because of the other problem this post addresses, they would not cause systemic death of commensal bacteria, just the ones they target.

The latest review on the subject is linked in my profile and should be still be open access for a while longer.
posted by Blasdelb at 8:36 PM on November 25, 2011 [1 favorite]

Unfortunately, a patient who has gotten antibiotics once for bronchitis will forever think that they need more of the same if bronchitis strikes again and will look askance at any doc who tries to convince them otherwise. One of my biggest professional pet peeves...

From the patient point of view I see what you mean. I'm hesitant to accept prescriptions for antibiotics and I get the impression that I'm an outlier based on the reaction from my Doctor. He agrees, but he confides that many people aren't satisfied with their visit if they don't leave with a prescription. I go to seek advice and consider it just fine if I'm advised to wait and see or try to alter my diet or any number of other cause seeking activities.
posted by dgran at 6:36 AM on November 28, 2011

The vast majority of sinus infections and bronchial infections are viral and not responsive to antibacterial drugs. You may not fit that pattern (although I doubt it unless you are prone to secondary bacterial infections pursuant to the viral infections that cause the vast majority of upper respiratory acute infections), but there is absolutely no question that doctors over-prescribe antibiotics to satisfy the demands of patients who are ignorant of the science involved.

"I doubt it" is the sort of thing I find inappropriate from someone who doesn't know my health history. Perhaps my doctors are prescribing me antibiotics fecklessly after I've had an allergic reaction to one that sent me to the ER, but I'm holding on to the illusion that they've read my chart.

rumposinc's excellent comment really gets to the heart of one problem: the health care system in the US is structurally unsound in a way that fairly directly leads to overmedication, and not just with antibiotics. Patients come in and want care, and because of the time and financial constraints in the system, doctors don't have the time to deliver care beyond immediate problems, and that usually in the form of a prescription, even when a prescription may not be the most appropriate treatment. Plus, for patients, there's the choice between trying to educate yourself and being painted as a problem for having ideas about how you should be treated (drug seeker on the one hand, refuses to medicate problems instead of attempting lifestyle changes on the other) or coming in ignorant, the problems with which have been discussed already. When you have any kind of respiratory illness, it's hard to find a doctor who'll treat you beyond mild poking and prodding and sending you off with a prescription--or without in some cases, with the operating assumption you're a drug seeker looking for a hit of antibiotics--when what you're really looking for is to know whether you're seriously ill or likely to become so, and how to best treat whatever you've got (preferably without compromising any other health issues).
posted by immlass at 8:59 AM on November 28, 2011 [1 favorite]