General Concepts Concerning the Use of Antimicrobials and Infectious Diseases

1) There is nothing intrinsically 'powerful', 'strong', or 'big gun' about an antibiotic. The use of these terms is a reliable marker (sensitivity and specificity of 99% in my hands) that the speaker knows nothing about antibiotics. They are marketing concepts, no more helpful than the phrase ‘new and improved’. I think, without data in this instance, that antibiotics are not used as chemicals to kill organisms in a specific body space. Rather, they are magical talismans that are used to ward off evil spirits. Instead of possession by demons, we have infection by Pseudomonas and instead of exorcism, we give Zosyn. The words may have changed, but the thinking behind it remains unchanged.

2) There is nothing intrinsically good about 'broad-spectrum', the concept of which, like the terms mentioned in 1, may lead to a false sense of security that you are appropriately treating the patients infection. The good thing about broad-spectrum antibiotics is that they breed really resistant infections, which in turn leads to ID consultations, which in turn pays my mortgage. On second thought, maybe broad-spectrum antibiotics are not such a bad idea after all..........

3) Most antibiotics work most of the time in most patients. That is why 40% of the time we give antibiotics inappropriately in the in hospital, yet patients do just fine. Most of the time. There are diseases (sepsis, pneumonia, and endocarditis to give three examples) where giving the wrong antibiotics actually leads to a worse outcome. Such as death, Really. Giving an antibiotic to which the organism is resistant can be a bad idea. People can and do die if you choose wrong. And since 40% of the time antibiotics are given incorrectly, it is important that the hospital formulary reflect this. The antibiotics available for people who do not know what they are doing should be those that can do the least harm. The formulary needs to protect patients not from the good doctors, but from those whose practice may be, shall we say, suboptimal.

4) Anyone who gets their information from a pharmaceutical rep is a self-deluded fool. Advertising works. Most doctors say they pay no attention to what drug reps say. But the data suggests that not only is the information one gets from the pharmaceutical rep wrong but it is their primary source of information on drugs. For the record, I have not talked to a rep, except for a terse hello, in over 15 years, and, with one exception (a rep sent me a Fleets enema. I have it proudly in my desk), have accepted nothing from industry. Unlike most of my colleagues, I have not whored myself. (But it is not to say I wouldn't. Given enough zeroes, my honor and integrity are for sale). Go to for the skinny.

5) I practice the "My Mothers Infected Prosthetic Valve Has To Be Treated Medically" or MMIPVHTBTM approach to infectious diseases. If antibiotic X is the best drug for MMIPVHTBTM, then why would I use an inferior antibiotic in other circumstances? The answer is, of course, I wouldn't.

6) ‘I like X and have used it to good effect over the years.’ If X equals a condiment you put on your cheeseburger, then that is a reasonable assertion. If X is an antibiotic, that makes no rational sense. Such an approach ignores all the difficulties of human memory and is not an appropriate paradigm for the prescription of antibiotics (or any other drug). If however, you say I use X because the in vitro and in vivo data support its use, it makes sense. It would not be good criterion for choosing condiments for your cheeseburger. So if your rationale for using a drug would be equally applied to mustard and onions, perhaps you should rethink the reasoning used for said drug.

7) In treating patients, then, we want to give antibiotics that kill the organisms that will kill our patients, and we want reliable empiric killing (in serious illnesses, empiric susceptibility bordering on 100%) and we want to give antibiotics that have good levels in the body fluid(s) that are infected, that are minimally toxic, that do not breed resistance and that are inexpensive. We want the most bang for the buck (ID doctors are all closet military men).

8) It is amazing how many times I get a consult or hear about a patient where it is evident the treating physician has no clue about the significance of a culture or what the best antibiotic is. Sending a S. aureus bacteremia home on po clindamycin is a good example. No one would treat a cancer without an oncologist or deliver a baby without Ob help, but for some reason, people feel it is appropriate to screw up, er, I mean, take care of complex infections without ID input.

It is the classic example of the Dunning Kruger effect (Ref): "The Dunning–Kruger effect is a cognitive bias in which unskilled individuals suffer from illusory superiority, mistakenly rating their ability much higher than average. "

The ID doc knows more than you about infections. Pay attention to what they say. For any number of infections following ID recommendations will improve care and decrease cost (PubMed).

9) The three most dangerous words in medicine: In my experience. If a doc wants to use an antibiotic (or any therapy) because in their experience it works, run. That is why we have evidence, to know how to treat infections. Also, pay no attention to someone who 'likes' a given antibiotic. Treatments should not be chosen like a flavor of ice cream.

Are 'cidal' antibiotics better than static? The data suggests no (PubMed). That being said, for meningitis, endocarditis and the neutropenic (not evaluated in this study) I would still try to kill rather than stun the bacteria. Outcomes in ID depend on sone indefinable balance of source control, host immune system, virulence of the organism and the antibiotic. When the first two are lacking, you have to make it up with the last.

There was a nice article in JAMA on the 4 Moments of Antibiotic Decision. Take a moment and read the whole thing (PubMed) but for those of you with a short attention span, here they are:

1) Does this patient have an infection that requires antibiotics?

2) Have I ordered appropriate cultures before starting antibiotics? What empirical antibiotic therapy should I initiate?

3) A day or more has passed. Can I stop antibiotics? Can I narrow therapy? Can I change from intravenous to oral therapy?

4) What duration of antibiotic therapy is needed for this patient’s diagnosis?

And there are the classic Observations on spiraling empiricism: its causes, allure, and perils, with particular reference to antibiotic therapy (PubMed) which is still behind a paywall. Everyone should read it. Get it from your librarian but I discussed it at length over at Science-Based Medicine.

Their Fallacies in Antibiotic Therapy

  • Broader is better.
  • Failure to respond is failure to cover.
  • When in doubt, change or add another.
  • Sickness requires immediate treatment.
  • Response implies diagnosis.
  • Bigger disease, bigger drugs.
  • Bigger disease, newer drugs.
  • Antibiotics are non-toxic.

I will add my own.

  • Once started, an antibiotic cannot be stopped.
  • Once a class of antibiotic is started, you need to stay in class.
  • Gotta double cover (i.e. give two antibiotics) a particular organism.
  • The primary reason a particular antibiotic is given is “I like it.”

Curious Cases

Relevant links to my Medscape blog

Weird Rashes

Why is ID different?


Flying Pigs

There is much to be Thankful for


Rare Quasi-ethical Dilemma

Phone Call.

The Nevernever

No Problem

Premature Closure

A Quadruple Whinge



What is Worse than Drunk Driving?



Nailed It


The Worst Part of Being a Doctor

110 Years Ago

Everything is Going to Hell in a Hand Basket

Conflicts? Who can tell.

Rate au Vent


Here I Go Again

Let's Add A Few A's

Imaginary Beasts and Where Not To Find Them

Last update 11/30/19